Biological monitoring involving children exposed to mercury from a barometer in a private residence

G Model

ARTICLE IN PRESS

TOXLET-8681; No. of Pages 9

Toxicology Letters xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet

Biological monitoring involving children exposed to mercury from a barometer in a private residence Paul T.J. Scheepers a,b,∗ , Marieke van Ballegooij-Gevers a , Henk Jans a a b

Office for Health, Environment & Safety, Public Health Services Brabant/Zeeland, Tilburg, The Netherlands Department for Health Evidence, Radboudumc, Nijmegen, The Netherlands

h i g h l i g h t s • • • •

A spill from a broken barometer caused a mercury spill of a few milliliter. Removal of the spill using a vacuum cleaner caused a widely dispersed contamination. Blood analysis confirmed uptake of mercury in one adult and two children. A strategy for human biological monitoring was presented.

a r t i c l e

i n f o

Article history: Received 29 January 2014 Received in revised form 21 March 2014 Accepted 23 March 2014 Available online xxx Keywords: Elemental mercury Chemical incident Inhalation exposure Decontamination Blood

a b s t r a c t A small spill of approximately 3 mL of mercury from a broken barometer in a residential setting resulted in blood values of 32 ␮g/L in a boy of 9 months and 26 ␮g/L in a girl of 2.5 years in samples collected within 6 h after the start of the incident. A nanny who attempted to remove the spill had a blood mercury value of 20 ␮g/L at the same time point. These elevated blood values were attributed to inhalation rather than dermal uptake or ingestion. Exposure was aggravated by the use of a vacuum cleaner in an early attempt to remove the spill and incomplete decontamination of involved persons, leading to a continuation of exposure. Over a period of three months general cleaning was followed by targeted cleaning of hot spots until the indoor air mercury levels reached a median value of 0.090 ␮g/m3 with a range of 0.032–0.140 ␮g/m3 . Meanwhile the family was staying in a shelter home. Human biological monitoring (HBM) was motivated by the complex exposure situation and the involvement of young children. Initially high blood values triggered alertness for clinical signs of intoxication, that (as it turned out) were not observed in any of the exposed individuals. Despite continued exposure from hair and clothes, within six weeks after the incident, blood levels returned to a background level normally seen in children. HBM contributed to reassurance of the parents of the young children that quick elimination of the mercury did not require medical treatment. © 2014 Published by Elsevier Ireland Ltd.

1. Introduction In many residences there are potential sources of exposure to elemental mercury. Private residences made up 16.7% of 406 elemental mercury spills recorded in 14 states in the US over a period of 6 years (Zeitz et al., 2002). In homes mercury can evaporate from broken objects such as antique or replica silvered mirrors, lamps

∗ Corresponding author at: Radboudumc, Department for Health Evidence (133), PO Box 9101, NL-6500 HB Nijmegen, The Netherlands. Tel.: +31 24 3616878; fax: +31 24 3613505. E-mail addresses: [email protected], [email protected], [email protected] (P.T.J. Scheepers).

and vases, technical appliances such as electrical switches, mercury discharge lamps, gas regulators and measurement instruments such as thermometers, thermostats, barometers, blood-pressure gauges and pendulum clocks (CDC, 2007; Hryhorczuk et al., 2006). In the US 42% of the residential spills were related to dropped or spilled mercury containers. Some intoxications are described related to use of elemental mercury in cosmetics (Lauwerys et al., 1987; Li et al., 2000; Tang et al., 2013). Mercury is also used for spiritual purposes in certain Afro-Caribbean and Latino cultures (Baughman, 2006). Use of mercury in such rituals resulted in cases of intoxications involving children (Ozuah, 2000, 2001; Wu et al., 2013). In other cases of residential exposures the liquid mercury was brought into the home from an unattended industrial facility (Azziz-Baumgartner et al., 2007) or school (Schwartz et al.,

http://dx.doi.org/10.1016/j.toxlet.2014.03.017 0378-4274/© 2014 Published by Elsevier Ireland Ltd.

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model TOXLET-8681; No. of Pages 9

ARTICLE IN PRESS P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

2

1992; Risher et al., 2003). Sometimes the origin of the mercury remained unknown (Cherry et al., 2002). More serious cases resulted in fatalities involving children and were related to the melting of mercury on a kitchen stove (Campbell, 1948; Moutinho et al., 1981; Rowens et al., 1991). Elemental mercury is primarily taken up by inhalation due to an efficient absorption into the blood. The median (range) absorption was estimated to be 69% (57–73%) in a study involving nine human volunteers (Sandborgh-Englund et al., 1988) and 79% (75–84%) in four volunteers (calculated from data by Kudsk, 1965). A third study reported an average of 74% for five volunteers (Hursh et al., 1976) and a fourth more recent study reported an absorptions of 77 and 81% for two volunteers (Pogarev et al., 2002). Some clinical toxicology handbooks report higher values for inhalation exposure (Ellenhorn and Barceloux, 1988; Haddad and Winchester, 1990; Friberg et al., 1986; Baselt, 2004). Dermal absorption of elemental mercury vapor contributed only 2.2% of pulmonary uptake in human volunteers (Hursh et al., 1989) but for direct skin contact with the liquid phase there are no experimental data to support a quantitative estimate. The uptake from ingestion is estimated to be very low due to the limited absorption of mercury from the gastrointestinal tract (WHO, 2003; Sandborgh-Englund et al., 2004). Even when elemental mercury was ingested in suicide attempts, the patients survived and clinical symptoms of intoxication were explained by inhalation (Lech and Goszcz, 2006). Before redistribution into the tissues a considerable fraction is eliminated via exhalation, with half-lives of 13–25 h (Hursh et al., 1976). Cherian et al. (1978) studied toxicokinetics in human volunteers inhaling radioactive mercury and estimated the elimination corresponding to 7% of the dose within three days. Within 7 days 9.2% of the dose appears in feces and 2.4% in urine. In another study volunteers were exposed to 400 ␮g/m3 for 15 min. Elimination during light exercise resulted in loss of 7.5–12% during the first three days. Only 1% of the dose was excreted in urine after 3 days and 8–40% after 30 days (Sandborgh-Englund et al., 1988; Jonsson et al., 1999). Elemental mercury readily passes through the blood–brain and placenta barrier and becomes trapped as divalent mercury. It accumulates in the head, chest and kidneys with clearance half-lives of 21, 43 and 64 days (Hursh et al., 1976). These are also the primary targets were effects occur. Acute elemental mercury inhalation can lead to the following symptoms, usually within hours of exposure: cough, chills, fever and shortness of breath, metallic taste, dysphagia, salivation, weakness, headaches and visual disturbances. Complaints of the gastrointestinal tract include nausea, vomiting and diarrhea. In severe intoxications chest radiography reveals interstitial pneumonitis, emphysema, pneumothorax that may resolve or progress to respiratory failure or cardiac arrest and death (Rowens et al., 1991; Jung and Aaronson, 1980; Moutinho et al., 1981). In many cases the respiratory distress that followed the initial symptoms is delayed for several days to a week (Jung and Aaronson, 1980; Moutinho et al., 1981). This study describes an acute exposure of one adult and two young children who were involved in a residential mercury spill from a broken barometer. In the follow-up human biological monitoring (HBM) was used in addition to extensive air monitoring. The results are discussed and compared to other reported cases of mercury spills in private residences. 2. Methods Air monitoring was used to assess the situation in terms of health risk and also to support effective cleaning of the house and furnishing. Mercury vapors were measured in a heated home (19–22 ◦ C) with ventilation that was close to the normal situation during the cold season (all windows and doors closed but some of the ventilation grids opened). Mercury vapor measurements were performed using a Lumex mercury spectrometer type RA-915M (Lumex Instruments, Nicosia, Cyprus).

Fig. 1. Plan of the residence. Numbers indicate concentrations of mercury (in ␮g/m3 ) measured at close range to floors and objects on December 6th (3 days after the incident). Numbers in circles indicate results of measurements performed at breathing height.

Two sampling strategies were used: building materials and objects were scanned at close range to search for hot spots of mercury evaporation. In a second sampling approach the air concentration at breathing height was determined (approximately 100 cm from the floor) central in the room. For the timeline day numbers were assigned, indicating day 1 as the day when the spill occurred. For the calculations of air concentrations multiple samples at the same location were combined by taking the average (to avoid the increase of the weight of such locations in the overall descriptive statistics). The parents approved the use of information from the medical files of the children. The nanny also approved use of her medical files. From the nanny and the children on the day of the incident a venous blood sample was collected and sent to the clinical laboratory of a hospital. In addition to mercury, also standard blood parameters were determined (including liver and kidney function). Blood collection was repeated on day 4, 39 and 122, following the day of the incident. Urine was collected on two occasions.

3. Results 3.1. Case description On Monday, December 3rd, 2012 around 14:00 h the Office for Health, Environment & Safety, the Public Health Services Brabant/Zeeland in Tilburg, The Netherlands (public health service) received a phone call related to a mercury spill, involving young children. The caller was the nanny, a 52-years-old woman, who was taking care of two children on behalf of the parents who were at work. The incident involved a boy of 9 months and a girl of 2.5-years-of-age. The incident which caused the contamination occurred at around 10:00 h in the morning in the hallway of a private residence, a modern house built in 1996 with concrete floors. In Fig. 1 the plan of ground floor and second floor of the residence is presented. The mercury was released from the glass reservoir of a traditional barometer produced in The Netherlands. It is estimated that 30% of the volume of liquid mercury was spilled. This corresponds to a volume of approximately 3 mL. The nanny and the children were putting on their coats in the hallway to go out when the boy touched the glass of a barometer on the wall causing it to break and mercury to spill. Mercury was spilled on the boy’s head, clothes and on the stroller and also on the hardwood floor. The stroller was contaminated with a stain of approximately 10 × 5 mm. The nanny put the boy on a rug on the floor and the girl in her own play area in the living room. In the meantime she cleaned the spill using a vacuum cleaner. The door between the hallway and the living room was left open for the nanny to keep an eye on the children. After cleaning she took the children outside for a walk. Upon return, several hours after the spill, the children were put to bed in their rooms on the first floor. Despite the indications

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model TOXLET-8681; No. of Pages 9

ARTICLE IN PRESS P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

of direct contact, personal whole body decontamination was not considered/attempted. Only the children’s clothes were changed. When the children were asleep the nanny went downstairs for more thorough cleaning. Following an instruction that she found on the Internet, she used the light of a torch to find additional mercury contamination. She removed the beads using adhesive tape. No gloves were used to protect her skin from direct contact with liquid and vapor. During this initial cleaning attempt shoes were not changed or cleaned. Some additional mercury beads were discovered on the wooden floor in the hallway and on the rug in the living room where the boy had been playing. The nanny read on the Internet about potential health risks related to mercury spills. Concerned with this information she decided to contact the public health service at 16:00. The broken barometer and the stroller were taken outside immediately when the father arrived at 16:30. When the mother came home later that afternoon she walked through the contaminated area to other parts of the house. After 15 min she put her contaminated footwear outside. After taking down the children, after their nap, fresh clothes were put on but the children walked through contaminated rooms before they were evacuated to their grandparent’s home. The nanny was still wearing the same (presumably contaminated) clothes and shoes at that time. When the nanny went home she changed clothes and shoes before entering her house. 3.2. Initial appraisal and advice The environmental physician of the public health service advised the parents to evacuate their home to allow ventilation and cleaning and return only after the air quality was evaluated for the level of mercury vapor. The nanny and the parents were advised to remove contaminated items such as the barometer, stroller, vacuum cleaner, rug of the play area, bed linen, clothes and materials used in the initial cleaning of the spill, put them in plastic bags and take them outdoors for later removal as chemical waste. The windows and doors were opened to increase ventilation. Two days later, a specialized chemical cleaning company removed all potentially contaminated small items. Larger items such as a couch, chairs and the curtains of the living room were removed several days/weeks later. Some of this furniture could not be cleaned, remained a source of mercury evaporation and were therefore discarded. On the second day following the incident, the environmental physician contacted the general practitioner to discuss the accident and the required follow-up health surveillance. The use of repeated blood collection was recommended to assess the systemic internal dose resulting from the exposure. This step was taken because of difficulty to assess the complex exposure situation and the involvement of young children. The parents were advised to take their children to the outpatient clinic for the collection of blood samples. A similar advice was given to the nanny. The first blood samples were taken approximately 6 h following the onset of the spill. 3.3. Assessment of contamination and forthcoming exposure Experts from the National Institute of Public Health and the Environment (RIVM) and from the public health service arrived on day 4 to assess the exposure situation and perform measurements in the residence. They noted that not all contaminated items had been removed by the cleaning company. Table 1 and Fig. 2 provide an overview of the results of measurements that were taken at close range from floors and furniture. The first measurements were focused on the situation in the hallway and the living room. Hot spots of mercury vapors were found in the hallway on the floor close to the baseboard below the position where the barometer had been hanging on the wall. Traces of mercury were also found

3

in seams between tiles in the kitchen, on the living room floor in front of the couch and on the floor in the play area where the rug had been removed. As the children were taken to bed the staircase to the first floor and the boy’s and girl’s bedrooms also became contaminated. The playground in the living room, the floor of the kitchen, and the stairs to the first floor were identified as sources of mercury evaporation. The median and range of measurements at close range were 7.0 (0.5 to 28) ␮g/m3 . At breathing height median values and ranges were 2.3 (0.7–12) ␮g/m3 (see Table 1). Afterwards, the professional cleaning company proceeded with decontamination of the house. They paid special attention to hot spots highlighted by measurements of the RIVM. Careful follow-up measurements on day 9 by the RIVM revealed that the contamination was spread and still present. Several different hot spots of persistent mercury evaporation were discovered on the floor in the hallway. Further hot spots were also found in the living room and the kitchen. The second evaluation measurements at close range identified some new hot spots of mercury evaporation in the bathroom on the 1st floor and on the car floor mat and child car seat. Even after removing textile carpet upholstering of the staircase to the first floor and to the second floor, hot spots of mercury evaporation were discovered by the RIVM. These sources of contamination resulted in a median air concentrations of 5.0 ␮g/m3 more than a week following the incident (Fig. 2). At this point, the public health service and RIVM advised the professional cleaning company to use a mobile mercury vapor monitor to optimize the cleaning process. The median air concentration at close range and at breathing height on day 9 were very similar to the values observed on day 4 (Table 1). The pattern of mercury levels at breathing height was much different compared to the pattern observed on day 4 (see Fig. 3). After further professional cleaning, new measurements were taken by the RIVM and public health service two and three months (day 64 and 86) after the spill occurred. These measurements were carried out to assess the possibility of safe return of the family to their home. As indicated in Table 1 and Figs. 2 and 3 the extended ventilation and cleaning resulted in overall very low values. After three months a median indoor air mercury concentration was 0.090 ␮g/m3 with a range of 0.032–0.140 ␮g/m3 and a 0.95 percentile value of 0.14 ␮g/m3 . Based on these values that were below the established guidance for indoor air quality of 0.2 ␮g/m3 for long-term inhalation exposure to elemental mercury vapor (WHO, 2003) it was decided that the family could be advised to return to their home. Although the exact time of evacuation was not recorded it is estimated that the nanny and the children had been exposed at the incident location during 4 h. When leaving for the shelter home personal decontamination had not been considered/attempted. Clothes were changed but not shoes. This presumably caused a secondary contamination that was discovered on day 9 (Table 2). Hair of the boy, and clothes the boy was wearing at that moment were secondary sources of mercury evaporation. The transfer of mercury by clothes and body surface contamination residues was confirmed by observed elevated mercury levels in the car interior and identification of the car seats of the children as sources of mercury evaporation (see Table 3). The tracking of mercury contamination with footwear was supported by the finding of sources of mercury vapor on the floor of the car in front of the driver’s seat. Footwear of the children were no source. This resulted in the continued exposure observed in the shelter home and confirms extended inhalation exposure during the period of evacuation, although at a much lower level (<0.2 ␮g/m3 ) than on the day of the incident. On day 9 this resulted in air concentrations of up to tenfold above outdoor background levels of mercury (see Tables 2 and 3). It is expected that these levels gradually decreased over the course of

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model TOXLET-8681; No. of Pages 9 4

ARTICLE IN PRESS P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

Fig. 2. Air concentration of mercury vapors at close range in ␮g/m3 .

Fig. 3. Air concentrations at breathing height in ␮g/m3 .

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model

ARTICLE IN PRESS

TOXLET-8681; No. of Pages 9

P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

5

Table 1 Median and range of air concentrations of mercury (␮g/m3 ) evaporating from surfaces and at breathing height. Type of measurement

Date Day number

Dec–06–2012 4

Dec–11–2012 9

Feb–07–2013 64

Mar–01–2013 86

Close range

Number of measurements Median (range) 95th percentile Number of measurements Median (range) 95th percentile

23 7.0 (0.50–28.0) 26.6 14 2.3 (0.70–12) 8.4

21 5.0 (0.06–21.5) 12.9 11 3.4 (1.5–6.4) 6.1

25 0.20 (0.095–0.70) 0.57 11 0.13 (0.040–0.30) 0.27

20 0.11 (0.025–0.38) 0.35 9 0.090 (0.032–0.14) 0.14

At breathing height

Table 2 Concentrations of mercury vapor in the shelter home were parents stayed with their children for 3 months, as measured on December 11th, 8 days after arrival (␮g/m3 ). Room

Source measurement

At appr. 100 cm from floor

Outdoor background Couch Chairs in living room Chairs dinner room Chair (boy) Bed (boy) Chair (girl) Bed (girl) Bathroom Bedroom parents

0.11 0.078–0.100 0.08–0.10 0.12 0.032–0.10 0.13 0.022 0.050–0.075 –

0.010 0.066 – – – – – 0.029 – 0.038

Table 3 Air concentrations of mercury due to evaporation from clothes and hair of children measured at close range in ␮g/m3 in shelter home and car, 8 days following the day of the incident. Child

Source

Source measurements

Boy

Inside trouser Hair Pullover Shoes Car seat Shoes Cap in car seat Car seat

0.300 0.170 0.200 0.100a 0.200–1.10b 0.100a 1.00 0.130b

Girl

a b

(2003–2004) monitoring survey (1.1 ␮g/L). Even if the elimination was influenced by uptake in the shelter home this is consistent with the bimodal pattern of elimination of elemental mercury from the blood with half-lives of 1.2 and 10.5 days (Sandborgh-Englund et al., 1988). In addition to blood analysis, on January 11th a spot urine sample from the boy was tested for mercury, revealing a background excretion of 1.4 ␮g/L (1.5 ␮g/g creatinine). From the girl the volume of urine collected was insufficient for testing. On January 18th a spot urine sample from the nanny was tested for mercury and showed no elevated level of mercury excretion (7.0 ␮g/L, 5.9 ␮g/g creatinine). 4. Discussion As in most published case-reports also in the case described here there are many gaps in knowledge concerning the description of the exposure situation. Nonetheless, interpretation of the collected data of mercury values observed in air, blood and urine is possible to a certain extent, in conjunction with earlier reports from controlled exposures of volunteers and also of published reports of similar spill incidents. It is attempted to learn from this case and also derive some suggestions that may be used in the future, especially concerning the use of HBM. 4.1. Initial appraisal and cleaning

In same range as outdoor background. Measurement in outdoor air.

the stay in the shelter home but no further measurements were taken to confirm this. 3.4. Results of biomonitoring On the day of the incident blood was collected 6 h after the onset of the incident. As shown in Table 4 the first blood values were 32, 26 and 20 ␮g/L for the boy, the girl and the nanny, respectively. These values were elevated but much lower than values observed in previous cases that did show clinical symptoms of intoxication (Schwartz et al., 1992; Moutinho et al., 1981). These values were reduced by 25–50% of the initial values on day 4, to 3.0–3.5% (boy and girl) on day 39 and to 0.6% on day 122 (only boy). This showed blood levels returning to a background level normally seen in children. The levels on days 39 and day 122 were just below the 90-percentile of children of 6–11 years observed in the NHANES Table 4 Blood mercury values in ␮g/L on the day of the incident and in three follow-up measurements. Person

Dec–03–2012 Day 1

Dec–07–2012 Day 4

Jan–11–2013 Day 39

Apr–2–2013 Day 122

Boy Girl Nanny

32 26 20

8.9 5.3 11

1.1 0.80 –

0.2 – –

For initial assessment of the situation three days after the incident a direct reading instrument for detection of mercury vapors was used and 6 out of 25 measurements turned out to be higher than 10 ␮g/m3 . These measurements confirmed that the initial appraisal of the situation by experts of the public health services and their recommendation to evacuate the home was justified. While the family was sting in the home of the grandparents, on three separate occasions an evaluation of the residual contamination in their residence was performed to assess the effectiveness of cleaning. It is known from previous studies, that, as the general background air concentration of mercury is reduced, it becomes easier to pinpoint hot spots (Baughman, 2006). In this case, some hotspots were found on locations were the children were playing soon after the initial spill (the playground area in the living room). The shift from high to much lower mercury levels measured at breathing height indicates that the first and second cleaning effort was focused on rooms with obviously high contamination levels such as the hallway and living room. The second evaluation showed that the kitchen and bathroom had relatively higher levels than the hallway and living room, though not higher than the initial values in those locations. This might be caused by differences in ventilation attempting to remove mercury by evaporation or by a difference in residual mercury contamination. These residues were presumably caused by tracking from contaminated shoes. This was confirmed by the identification of the staircase as a hot spot and shows the added value of repeated air monitoring in conjunction with cleaning efforts.

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model

ARTICLE IN PRESS

TOXLET-8681; No. of Pages 9

P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

6

Table 5 Modeling of dermal absorption of mercury liquid and vapor. Duration of exposure (h)

Air concentration (␮g/m3 )

Dermal uptake by

% of total body skin surface

Uptake (␮g)

% of uptake by inhalation

Home

4

Shelter home

168

– 100 – 0.1

Liquid phase Vapor phase Liquid phase Vapor phase

1 100 0.01 100

1.7 0.38 0.0080 0.00018

0.2 0.06 0.6 0.001

The following properties of elemental mercury were used: vapor pressure of 0.2 Pa at 20 ◦ C; Water solubility of 0.2 ␮g/L at ◦ C and an octanol/water partition coefficient (Log Po/w ) of 4.5. Calculations using SkinperX (Wilschut et al., 1995; Patel et al., 2002; Ten Berge, 2009).

In addition to cleaning, removal of mercury by evaporation is effective to reduce indoor air mercury levels (Baughman, 2006). On the first day the home was ventilated by opening doors and windows. The cleaning was performed in the cold season with average outdoor temperatures of −0.8 to 7.3 ◦ C. After several weeks an additional industrial ventilator was placed to increase the air exchange rate. During two months following the day of the incident the indoor temperature was kept at 19 ◦ C (range: 15.7–22 ◦ C). During the third month and during the final evaluation of the air quality the temperature was raised to an average of 21–22 ◦ C. During the fourth and last visit by the RIVM and the public health service a residual contamination of mercury was discovered that had penetrated through the wooden top floor into the supporting concrete. This source of mercury evaporation could not be removed and was sealed by the chemical cleaning company using a gas-tight sealant coating. In the US 22% of residential spills resulted in evacuations of the occupants (Zeitz et al., 2002). The duration of the evacuation ranged from 4 h to 46 days. In the present case it took almost three months of alternating cleaning and air monitoring before the air concentrations at breathing height were below the WHO guidance for an indoor air level of mercury of 0.2 ␮g/m3 . This long period was probably due to the wide dispersion of contamination which required multiple cleaning efforts and also removal of wooden floors. The cold weather conditions may have further slowed down the evaporation of liquid residues (Baughman, 2006). The expenses of the cleaning of the residence amounted to approximately D 200,000. 4.2. Conditions explaining high inhalation exposure The exposure was aggravated by the use of a vacuum cleaner to remove the mercury spill in the initial cleaning attempt. Zelman et al. (1991) reported that a majority of adults questioned about removing a mercury spill reported that they would use a vacuum cleaner. Such use of a vacuum cleaner to remove the spill is known to cause wide dispersion of mercury (Baughman, 2006; Cherry et al., 2002; CDC, 2007). In this case a bagless vacuum cleaner was used. Mercury beads extracted from seams and cracks in the floor will break up into smaller aerosols and will also likely be evaporated due to the heat produced by the vacuum cleaner and be dispersed. The formation of ultrafine condensation nuclei increased the dispersion supported by air circulation in the residence (Schwartz et al., 1992; Hryhorczuk et al., 2006). When the air cools down it is likely that ultrafine mercury particles may soil surfaces (Baughman, 2006). A second factor contributing to wide dispersion was the incomplete decontamination of skin, hair and clothes. As shown in Table 3 measurements taken eight days following the day of the incident still showed elevated levels of mercury vapors from hair and clothes of the boy, up to 10-fold local outdoor background concentrations. The decontamination of the children was incomplete. On the day of the incident the clothes were changed but a whole body decontamination (shower or bath) was not performed. This occurred one or several days after the day of the incident. Eight days following the incident mercury was still evaporating from the boy’s hair. Apparently, a regular bathing routine in this case did not remove

all of this contamination. Following the continued evaporation of mercury from the boy’s hair the public health service advised a complete haircut to remove the residual contamination. In similar case studies hair contamination was reported to persist and lead to re-exposure (Baughman, 2006). A treatment of hair with shampoo containing selenium sulfide to bind metallic mercury was used in other cases where children had contaminated hair and scalps (CDC, 2005). Presumably the contaminated hair of the boy caused extended exposure to elevated air concentrations of mercury (see Table 2). In future incidents involving direct-contact a recommendation to remove all clothes and perform an appropriate decontamination should be mentioned in the first communication with the public health service. 4.3. Exposure assessment Visual inspection and air monitoring of mercury vapor showed that there were numerous sources of secondary contamination that likely caused finger shunt and ingestion of liquid mercury. This is only a minor route of uptake because absorption via the gastrointestinal tract is thought to be negligible (WHO, 2003). Skin absorption of vapor mercury does not lead to a significant contribution to internal levels (Hursh et al., 1989). For percutaneous absorption of liquid mercury quantitative data could not be retrieved from literature. However, there are indications from medicinal and cosmetic dermal applications of elemental mercury that direct contact with liquid mercury may be a route of uptake (Lauwerys et al., 1987; Tang et al., 2013). In the described case the dermal route may have been more important due to direct skin contact of the children with liquid mercury (contaminated skin, hair and clothes) and potentially also in the adult (unprotected cleaning). This was further studied by modeling the dermal absorption (Table 5). Two exposure scenarios were used: a scenario at the time of the spill, assuming exposure during 4 h to liquid mercury of 1% of the body surface in all of the persons and an average level of 100 ␮g/m3 of mercury vapor for the entire body surface. This scenario resulted in an estimated contribution of dermal uptake of 2.1 ␮g corresponding to 0.26% of the uptake by inhalation. The second scenario assumed an exposure at the shelter home (only relevant to the children) and assumed an exposure of 30 days of 0.01% of the body surface and an average 1000-fold lower air level (0.1 ␮g/m3 ). This scenario showed an uptake of 0.0082 ␮g, which represents approximately 0.60% of the uptake by inhalation. As these are worst case assumptions it may be assumed that the dermal uptake provides a minor contribution to total body uptake of the spilled elemental mercury. 4.4. Biomonitoring The use of HBM was justified, based on the complex exposure situation and the involvement of young children. HBM provided integration over different routes of exposure and repeated followup sampling provided useful data to assess the decrease in the blood values over time. HBM using blood analysis, provided valuable information on initial uptake at the incident location. The first blood samples were

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model

ARTICLE IN PRESS

TOXLET-8681; No. of Pages 9

P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

7

Table 6 Proposed urine and blood biomonitoring strategy following inhalation of mercury following an incident. Field of application

Urine Chronic exposure

Blood Peak exposure

Exhaled air Peak exposure

Half life

63.2 (12.8–98.9) days (Jonsson et al., 1999; Sandborgh-Englund et al., 1988) After 10 days for high exposures to 6 months for low exposures (Nakaaki et al., 1978)

1.2 and 10.5 days (Sandborgh-Englund et al., 1988) First sample to be taken within 24 h following the incident (Sandborgh-Englund et al., 1988)

First void in the morning due to diurnal variations (Piotrowski et al., 1975; Mason et al., 2001) Creatinine correction

Standardize time of the day

18 (13–25) h (Hursh et al., 1976) 17 h (Pogarev et al., 2012) In volunteer studies over a period of three days 7.0–12% of the dose was observed end-exhaled air (Jonsson et al., 1999; Hursh et al., 1976; Pogarev et al., 2002) Not critical. Repeated sampling is useful for modeling

Not applicable

Less sensitive to seafood and fresh dental amalgam (ACGIH, 2013) NHANES (http://wwwn.cdc.gov/nchs/ nhanes/search/nhanes03 04.aspx) (NHANES, 2003–2004)

Sensitive to seafood and fresh dental amalgam (ACGIH, 2013) Based on 98 studies of normal levels (Brune et al., 1991): 2.0 ␮g/L (mean) and 4.3 ␮g/L (90th percentile)

Time window for collection of samples

Time of day

Adjustments

Background Reference

collected within 6 h after the start of the incident. This is within the time-window of 0.3 to 24 h when peak exposures were observed in nine volunteers exposed by inhalation for 15 min to mercury levels of 400 ␮g/m3 (Sandborgh-Englund et al., 1988). HBM complemented the assessment of some of the health complaints that was difficult due to the young age of the children. Follow-up blood tests also contributed to reassurance of the parents of the children that quick elimination of the mercury occurred and no medical treatment was required. From these first series of blood values it can be seen that the children were initially somewhat higher internally exposed than the nanny (Table 4). The possibility that the exposure may potentially lead to a higher exposure than in adults should be considered. First of all, the young children have a breathing zone closer to the floor than adults, resulting in inhalation of higher air levels (mercury vapor is much heavier than ambient air). In this case the children had been playing on the floor at the time when the mercury spill was removed by vacuum cleaning. Second, children also have a somewhat higher tidal volume per kg of body weight compared to adults. The uptake in children may therefore be higher than in adults and lead to a relatively higher body burden. In the described case it is likely that inhalation from their own contamination of skin and clothes combined with their playing on the floor in the living room during the first cleaning attempt (use of a vacuum cleaner). In addition to having a higher exposure, children are known to be relatively more sensitive to neurotoxic substances (Counter and Buchanan, 2004; Bose-O’Reilly et al., 2008, 2010). An overview of the different biological media for HBM of mercury is given in Table 6. Plasma or blood values reflect the systemic available mercury that reaches the brain due to passage of mercury through the blood–brain barrier. Following a short-term exposure urine samples are not suitable, due to a latency of 10 days or more, caused by conversion of elemental mercury to the ionic bivalent analogue and subsequent binding to sulfhydryl groups in plasma proteins. In case studies of acute exposure urine mercury levels did not reflect recent uptake (Jung and Aaronson, 1980). In the current study a spot urine was collected on day 39 (boy) and day 26 (nanny), with mercury levels within the background range observed in the general population. This creatinine adjusted mercury excretion reconfirms that the exposure following evacuation and upon return to the home did not result in chronic exposure to high levels of mercury (ACGIH, 2013). If blood collection is not possible, the determination of mercury in exhaled air may instead be used as a non-invasive alternative for initial screening (Pogarev et al., 2002; Johnsson et al., 2005). Fresh

No applicable if collecting only the last 300–400 mL of an exhalation (end-exhaled air) Not sensitive to seafood. Sensitive to fresh dental amalgam (ACGIH, 2013) Pogarev et al. (2002) reported background values below 5 ng/m3 for two human volunteers

dental amalgam fillings may cause a background level in blood and exhaled air, unrelated to exposure from ambient sources. For blood samples the consumption of seafood causes enhanced levels of mercury, potentially masking other exposures. To avoid this, the patient can be instructed not to consume seafood 24 h prior to blood collection. 4.5. Similar cases resulting in health effects Early reporting and adequate response will limit health consequences. Spills that occur at home rarely lead to clinical intoxications (Baughman, 2006). Below, some of the more serious cases are described including a number of fatalities. They illustrate that often symptoms are initially not recognized as mercuryrelated, causing delay of appropriate treatment. In 1948, Campbell reported a fatality in a 4-month-old-infant who had been exposed to a small volume of mercury (described as one teaspoon) evaporated on a kitchen stove. Schwartz et al. (1992) reported on a case involving a family with two adults and two sons of 3.5 and 5 years old. A jar containing 2.3 kg liquid mercury fell from a closet in an apartment. In an attempt to clean up the spill a vacuum cleaner was used. Approximately two weeks following this incident a blood mercury level of 161 ␮g/L was observed in the 3.5-year-old boy and 172 ␮g/L in the 5-year-old boy one month following the incident. The father and mother had blood mercury levels of 131 and 125 ␮g/L, respectively. Both children were hospitalized, suffering from several symptoms such as fever, cough and total body rash. The 3.5-year-old child also had idiopathic thrombocytopenic purpura (ITP), a symptom that was not related to mercury exposure before. The parents did not have any clinical signs of intoxication except for frequent nosebleeds reported by the mother. These symptoms, including ITP, are also observed in viral diseases, but in both children all viral cultures turned out to be negative. Both children and their parents recovered upon treatment with chelation therapy and returned to their apartment after it was decontaminated. Another case describes a girl of 7 months of age, treated for an airway infection (Moutinho et al., 1981). Later on, it became clear that she had been crawling on a kitchen floor while her father was melting a lead containing metal alloy with a high content of mercury. On the fourth day of hospitalization a blood mercury value of 35 ␮g/L was observed. On the seventh day she became dyspnoic leading to apnea. A bilateral pneumothorax with complete collapse of both lungs was observed. Despite treatment she developed acidosis, convulsions and coma and died 6 h later. Another fatality involved a 53-year-old man who had been destillating ore

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model TOXLET-8681; No. of Pages 9

ARTICLE IN PRESS P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

8

using elemental mercury on his kitchen stove for about 20 min. He was admitted to hospital with chills, fever, dyspnea, headache, diarrhea and abdominal cramps. When he was admitted to hospital four days after a second exposure to mercury a roentgenogram of the patient’s chest showed bilateral lung infiltrates consistent with pulmonary edema and massive pneumothorax followed by empyema. During the third week of hospital admission the patient acquired nosocomial pneumonia and died (Jung and Aaronson, 1980). In a similar case of mercury contamination of a home, chronic exposures due to inadequate response resulted in clinical signs in seven family members, including a 3-year-old girl who suffered from progressive weight loss, limping, ataxia, irritability, ‘screaming’ and regression in speech capability (Cherry et al., 2002). In the case described in this paper it is likely that the early evacuation on the day of the incident was instrumental in prevention of a continued high exposure associated with the potential development of clinical symptoms. 5. Conclusion Following a spill of a very small amount of elemental mercury in a residence, biomonitoring was applied to evaluate total body burden due to different sources and routes of uptake. The initial high mercury blood level triggered alertness for acute clinical signs of intoxication. The mercury blood levels were of concern but required no further medical treatment. There were no signs of mercury intoxication and follow-up samples showed a rapid decline in blood mercury. Early evacuation to a shelter home contributed to the return within 6 weeks of the blood values to levels normally seen in children. Follow-up blood sampling contributed to reassurance of the parents that the toxic substance was eliminated from the body of their children. This case report showed that the use of a vacuum cleaner to remove a 3 mL spill of mercury in a residence can lead to a complex exposure setting with significantly elevated blood mercury values. Exposure was extended over time due to initial incomplete personal decontamination. Cleaning of the residence took three months. The cleaning effort and final evaluation of the air quality was supported by air monitoring by an independent health authority. Early reporting of a spill to local or regional health authorities will result in information concerning health risks and advice to minimize the exposure by early evacuation, complete personal decontamination, and a measurement-supported cleaning and decision for a safe return of the residents to their home. In future events this will likely prevent uptake of mercury by occupants such as described in this case report. Conflict of Interest The authors declare that there are no conflicts of interest. Transparency document The Transparency document associated with this article can be found in the online version. Acknowledgments The authors acknowledge the parents and the nanny, the general practitioner, the clinical chemist, the National Poison Information Center (NVIC), technicians of the National Institute of Public Health and the Environment (RIVM) who performed the mercury vapor measurements. References ACGIH, 2013. Biological exposure index (BEI) documentation. In: American Conference of Governmental Industrial Hygienists, Cincinnati, OH.

Azziz-Baumgartner, E., Luber, G., Schurz-Rogers, H., Backer, L., Belson, M., Kieszak, S., Caldwell, K., Lee, B., Jones, R., Todd, R., Rubin, C., 2007. Exposure assessment of a mercury spill in a Nevada school—2004. Clin. Toxicol. (Phila.) 45, 391–395. Baselt, R.C., 2004. Mercury: Disposition of Toxic Drugs and Chemicals in Man, seventh ed. Biomedical Publications, Foster City, CA, pp. 668–671. Baughman, T.A., 2006. Elemental mercury spills. Environ. Health Perspect. 114, 147–152. Bose-O’Reilly, S., Lettmeier, B., Gothe, R.M., Beinhoff, C., Siebert, U., Drasch, G., 2008. Mercury as a serious health hazard for children in gold mining areas. Environ. Res. 107, 89–97. Bose-O’Reilly, S., McCarty, K.M., Steckling, N., Lettmeier, B., 2010. Mercury exposure and children’s health. Curr. Probl. Pediatr. Adolesc. Health Care 40, 186–215. Brune, D., Nordberg, G.F., Vesterberg, O., Gerhardsson, L., Wester, P.O., 1991. A review of normal concentrations of mercury in human blood. Sci. Total Environ. 100, 235–282. Campbell, J.S., 1948. Acute mercurial poisoning by inhalation of metallic vapour in an infant. Can. Med. Assoc. J. 58, 72–75. Centers for Disease Control and Prevention (CDC), 2005. Measuring exposure to an elemental mercury spill—Dakota County, Minnesota, 2004. MMWR Morb. Mortal. Wkly. Rep. 54, 146–149. Centers for Disease Control and Prevention (CDC), 2007. Elemental mercury releases attributed to antiques—New York, 2000–2006. MMWR Morb. Mortal. Wkly. Rep. 56, 576–579. Cherian, M.G., Hursh, J.B., Clarkson, T.W., Allen, J., 1978. Radioactive mercury distribution in biological fluids and excretion in human subjects after inhalation of mercury vapor. Arch. Environ. Health 33, 109–114. Cherry, D., Lowry, L., Velez, L., Cotrell, C., Keyes, D.C., 2002. Elemental mercury poisoning in a family of seven. Fam. Publ. Health 24, 1–8. Counter, S.A., Buchanan, L.H., 2004. Mercury exposure in children: a review. Toxicol. Appl. Pharmacol. 198, 209–230. Ellenhorn, M.J., Barceloux, D.G., 1988. Medical Toxicology, Diagnosis and Treatment of Human Poisoning. Elsevier, Amsterdam, pp. 1048–1053. Friberg, L., Nordberg, G.F., Kessler, E., Vouk, V.B., 1986. Metabolism and toxic effect of elemental mercury and inorganic mercury compounds. In: Handbook of the Toxicology of Metals. Elsevier Science Publishers, Amsterdam, The Netherland, pp. 396–409. Haddad, L.M., Winchester, J.F., 1990. Clinical Management of Poisoning and Drug Overdose. WB Saunders, Philadelphia, PA, pp. 1002–1009. Hryhorczuk, D., Persky, V., Piorkowski, J., Davis, J., Moomey, C.M., Krantz, A., Runkle, K.D., Saxer, T., Baughman, T., McCann, K., 2006. Residential mercury spills from gas regulators. Environ. Health Perspect. 114, 848–852. Hursh, J.B., Cherian, M.G., Clarkson, T.W., Vostal, J.J., Mallie, R.V., 1976. Clearance of mercury (HG-197, HG-203) vapor inhaled by human subjects. Arch Environ Health. 31 (6), 302–309. Hursh, J.B., Clarkson, T.W., Miles, E.F., Goldsmith, L.A., 1989. Percutaneous absorption of mercury vapor by man. Arch. Environ. Health 44, 120–127. Johnsson, C., Schütz, A., Sällsten, G., 2005. Impact of consumption of freshwater fish on mercury levels in hair, blood, urine, and alveolar air. J. Toxicol. Environ. Health A 68, 129–140. Jonsson, F., Sandborgh-Englund, G., Johanson, G., 1999. A compartmental model for the kinetics of mercury vapor in humans. Toxicol. Appl. Pharmacol. 155, 161–168. Jung, R.C., Aaronson, J., 1980. Death following inhalation of mercury vapor at home. West J. Med. 132, 539–543. Kudsk, F.N., 1965. Absorption of mercury vapour from the respiratory tract in man. Acta Pharmacol. Toxicol. (Copenh.) 23, 250–262. Lauwerys, R., Bonnier, C., Evrard, P., Gennart, J.P., Bernard, A., 1987. Prenatal and early postnatal intoxication by inorganic mercury resulting from the maternal use of mercury containing soap. Hum. Toxicol. 6, 253–256. Lech, T., Goszcz, H., 2006. Poisoning from aspiration of elemental mercury. Clin. Toxicol. (Phila.) 44, 333–336. Li, A.M., Chan, M.H., Leung, T.F., Cheung, R.C., Lam, C.W., Fok, T.F., 2000. Mercury intoxication presenting with tics. Arch. Dis. Child 83, 174–175. Mason, H.J., Hindell, P., Williams, N.R., 2001. Biological monitoring and exposure to mercury. Occup Med (Lond) 51, 2–11. Moutinho, M.E., Tompkins, A.L., Rowland, T.W., Banson, B.B., Jackson, A.H., 1981. Acute mercury vapor poisoning: fatality in an infant. Am. J. Dis. Child 135, 42–44. Nakaaki, K., Fukabori, S., Tjada, O., 1978. On the evaluation of mercury exposure – A proposal of the standard value for health care of workers. J Sci Labour 54, 1–8. NHANES 2003-2004 http://wwwn.cdc.gov/nchs/nhanes/search/nhanes03 04.aspx (Last access January 24 2014). Ozuah, P.O., 2000. Mercury poisoning. Curr. Probl. Pediatr. 30, 91–99. Ozuah, P.O., 2001. Folk use of elemental mercury: a potential hazard for children? J. Nat. Med. Assoc. 93, 320–322. Patel, H., ten Berge, W., Cronin, M.T., 2002. Quantitative structure-activity relationships (QSARs) for the prediction of skin permeation of exogenous chemicals. Chemosphere 48, 603–613. Piotrowski, J.K., Trojanowska, B., Mogilnicka, E.M., 1975. Excretion kinetics and variability of urinary mercury in workers exposed to mercury vapour. Int Arch Occup Environ Health. 35, 245–246. Pogarev, S.E., Ryzhov, V., Mashyanov, N., Sholupov, S., Zharskaya, V., 2002. Direct measurement of the mercury content of exhaled air: a new approach for determination of the mercury dose received. Anal. Bioanal. Chem. 374, 1039–1044. Risher, J.F., Nickle, R.A., Amler, S.N., 2003. Elemental mercury poisoning in occupational and residential settings. Int. J. Hyg. Environ. Health 206, 371–379.

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017

G Model TOXLET-8681; No. of Pages 9

ARTICLE IN PRESS P.T.J. Scheepers et al. / Toxicology Letters xxx (2014) xxx–xxx

Rowens, B., Guerrero-Betancourt, D., Gottlieb, C.A., Boyes, R.J., Eichenhorn, M.S., 1991. Respiratory failure and death following acute inhalation of mercury vapor: a clinical and histologic perspective. Chest 99, 185–190. Sandborgh-Englund, G., Elinder, C.G., Johanson, G., Lind, B., Skare, I., Ekstrand, J., 1988. The absorption, blood levels and excretion of mercury after a single dose of mercury vapour in humans. Toxicol. Appl. Pharmacol. 150, 146–153. Sandborgh-Englund, G., Einarsson, C., Sandström, M., Ekstrand, J., 2004. Gastrointestinal absorption of metallic mercury. Arch. Environ. Health 59, 449–454. Schwartz, J.G., Snider, T.E., Montiel, M.M., 1992. Toxicity of a family from vacuumed mercury. Am. J. Emerg. Med. 10, 258–261. Tang, H.L., Mak, Y.F., Chu, K.H., Lee, W., Fung, S.K., Chan, T.Y., Tong, K.L., 2013. Minimal change disease caused by exposure to mercury-containing skin lightening cream: a report of 4 cases. Clin. Nephrol. 79, 326–329. Ten Berge, W., 2009. A simple dermal absorption model: derivation and application. Chemosphere. 75, 1440–1445.

9

WHO, 2003. Elemental mercury and inorganic mercury compounds: human health aspects. In: Concise International Chemical Assessment Document 50. WHO, Geneva, http://www.who.int/ipcs/publications/cicad/en/cicad50.pdf (last access January 24th, 2014). Wilschut, A., ten Berge, W.F., Robinson, P.J., McKone, T.E., 1995. Estimating skin permeation: the validation of five mathematical skin permeation models. Chemosphere 30, 1275–1296. Wu, M.L., Deng, J.F., Lin, K.P., Tsai, W.J., 2013. Lead, mercury, and arsenic poisoning due to topical use of traditional Chinese medicines. Am. J. Med. 126, 451–454. Zeitz, P., Orr, M.F., Kaye, W.E., 2002. Public health consequences of mercury spills: Hazardous Substances Emergency Events Surveillance system, 1993–1998. Environ Health Perspect. 110, 129–132. Zelman, M., Camfield, P., Moss, M., Camfield, C., Sweet, L., 1991. Toxicity from vacuumed mercury: a household hazard. Clin. Pediatr. (Phila.) 30, 121–123.

Please cite this article in press as: Scheepers, P.T.J., et al., Biological monitoring involving children exposed to mercury from a barometer in a private residence. Toxicol. Lett. (2014), http://dx.doi.org/10.1016/j.toxlet.2014.03.017