Urinary excretion of platinum from South African precious metals refinery workers
=================================================================================

* Stephanus J L Linde
* Anja Franken
* Johannes L du Plessis

## Abstract

**Background** Urinary platinum (Pt) excretion is a reliable biomarker for occupational Pt exposure and has been previously reported for precious metals refinery workers in Europe but not for South Africa, the world’s largest producer of Pt.

**Objective** This study aimed to quantify the urinary Pt excretion of South African precious metals refinery workers.

**Methods** Spot urine samples were collected from 40 workers (directly and indirectly exposed to Pt) at two South African precious metals refineries on three consecutive mornings prior to their shifts. Urine samples were analysed for Pt using inductively coupled plasma-mass spectrometry and were corrected for creatinine content.

**Results** The urinary Pt excretion of workers did not differ significantly between sampling days. Urinary Pt excretions ranged from <0.1 to 3.0 µg Pt/g creatinine with a geometric mean of 0.21 µg Pt/g creatinine (95% CI 0.17 to 0.26 µg Pt/g creatinine). The work area (P=0.0006; η2=0.567) and the number of years workers were employed at the refineries (P=0.003; η2=0.261) influenced their urinary Pt excretion according to effect size analyses. Directly exposed workers had significantly higher urinary Pt excretion compared with indirectly exposed workers (P=0.007).

**Conclusion** The urinary Pt excretion of South African precious metals refinery workers reported in this study is comparable with that of seven other studies conducted in precious metals refineries and automotive catalyst plants in Europe. The Pt body burden of workers is predominantly determined by their work area, years of employment in the refineries and whether they are directly or indirectly exposed to Pt.

*   biomonitoring
*   exposure assessment
*   body burden
*   platinum group metals

### Key messages

#### What is already known about this subject?

*   Urinary platinum (Pt) excretions of precious metals refinery workers have been reported for Europe but not for South Africa, the largest producer of Pt in the world.

#### What are the new findings?

*   This is the first paper to report the urinary Pt excretions of South African precious metals refinery workers. Their urinary Pt excretions are comparable with concentrations reported for precious metals refinery workers in Europe.

*   The urinary Pt excretions of workers were determined by their work area, years of employment in the refineries and whether they had direct or indirect contact with Pt compounds.

#### How might this impact on policy or clinical practice in the foreseeable future?

*   Urinary Pt excretions can be used for risk assessment and to indicate increased exposure or Pt body burden in South African precious metals refineries.

## Introduction

Allergic reactions of the airways and skin resulting from occupational exposure to soluble platinum (Pt) compounds are well known and have been frequently reported in precious metals refinery workers.1–5 Cases of refinery workers suffering from asthmatic symptoms caused by soluble Pt compounds were first reported in British refineries in 1945 where the symptoms were linked with respiratory exposure.1 Since then, a number of investigations have associated sensitisation to soluble Pt with the degree of exposure to soluble Pt compounds experienced by workers,2 5 6 and increased urinary Pt excretions have been observed in workers who work in high exposure areas.7 Biological monitoring techniques such as the assessment of urinary Pt excretion can, therefore, be used to assess the exposure experienced by workers.8 In order to accurately assess exposure to a chemical substance through biological monitoring, the kinetics of the chemical in the human body needs to be understood since it affects the biological matrix that is used and the timing of the sample.9

The excretion of Pt is slow.10 11 Schierl *et al*11 exposed two human volunteers to soluble Pt dust through the handling of dry ammonium hexachloroplatinate (NH4)2PtCl6) powder for 4 hours. The urinary Pt excretion from these volunteers reached a maximum approximately 10 hours following cessation of exposure and followed a biphasic exponential decay pattern, which corresponded to a first half-life of 50 hours (95% CI 36 to 66 hours) and a second half-life of 24 days (95% CI 18 to 33 days). Even 166 days after exposure had ceased, the excretion of Pt via the urine was still above the baseline concentrations. Schierl *et al*11 only stated inhalation as a route of exposure, and it is unclear whether the volunteers were exposed via the skin and/or ingestion pathways. This is important since in vitro permeation studies have reported that Pt (from hexachloroplatinate salts) may permeate through intact human skin and that the skin may possibly serve as an exposure route for Pt.12 13 Schierl *et al*11 also reported that urinary Pt excretion from employees who had not been exposed to Pt for several years were still 25-fold higher than that of non-exposed subjects, and Weber *et al*10 reported no decrease in urinary Pt excretion from automotive catalyst production workers following 2 weeks of vacation. It was suggested that a long-term Pt reservoir could form in the body from where Pt may be gradually released for several years after exposure ends.11

Urine has been identified as a reliable biological matrix for use when performing biological monitoring of Pt exposure and has been used in several studies in occupational settings.7 8 11 14 15 The concentration of Pt in urine may be used to distinguish between workers who work in high exposure areas and those who work in areas with lower exposure.7 16 Cristaudo *et al*8 reported Pt concentrations in workplace air and in biological samples (urine, blood and hair) of catalyst production and metal recycling workers. They observed a very strong positive correlation between Pt concentrations in the workplace air and Pt concentrations in the urine of workers and reported that urine is a reliable biomarker of short-term exposure to Pt. However, the Pt content of biological samples was not associated with hypersensitivity. Nevertheless, urinary Pt can be used to identify workers who have an increased Pt body burden and who, as a result of increased exposure, have an increased risk of becoming sensitised. Urinary Pt, therefore, indicates increased risk for sensitisation as a result of increased Pt intake.

In addition to refinery and catalyst production workers, urinary Pt excretion has been measured in workers who were occupationally exposed to Pt in roadside dust, hospital workers who prepared the antineoplastic drug, cisplatin, and dental technicians who treated dental alloys with Pt.17–21 However, these concentrations did not compare with the high concentrations reported for refineries and automotive catalyst production plants.4

Tighter restrictions on motor vehicle emissions have led to the increased production of automotive catalysts and an increase in the demand for Pt.8 22 This has led to an increased number of workers potentially being exposed to Pt compounds. Although urinary Pt excretions of workers have been reported for precious metals refineries in Europe, no concentrations have been reported for South Africa, the largest supplier of Pt in the world.11 15 22 Therefore, the main aim of this study was to quantify the urinary Pt excretion of South African precious metals refinery workers in order to compare it with the published literature. Additionally, since previous studies have demonstrated that urinary Pt excretions can be used to identify workers who experienced increased exposure to Pt,7 this study also aimed to distinguish between the urinary Pt excretions of different groups of workers (eg, workers from various work areas).

## Methods

### Study population

Forty workers (32 men and 8 women; 31 African and 9 Caucasian) from two South African precious metals refineries were included in the study. Only workers employed at the refineries for longer than 1 year were included. The workers were aged between 22 and 56 years (mean=34.6±7.9 years), and the number of years which they were employed at the refinery was between 1 and 27 years (mean 7.7±6.3 years). The term *years of employment* refers to the time that workers worked in the refinery and not necessarily the time they worked at specific areas. As a control group, 10 persons (7 men and 3 women; 2 African and 8 Caucasian) aged between 25 and 62 (mean=36.5) who lived >100 km from the nearest Pt industry were included. For the purpose of this study the term *race* is used to define specific population groups based on genetic similarities, namely skin colour and physical features.23 The participants all received information regarding the details of the study prior to the start and provided written consent to participate in the study.

### Workplace description

Workers from various work areas within the refineries were included. These work areas included concentrate handling, platinum group metals (PGM) separation, crushing and ignition, precious metals, other precious metals, other production activities, other non-production activities, security, and the health clinic. In the concentrate handling area, the PGM concentrate was received, sieved and concentrated further, after which the concentrate was loaded into the reactors of the PGM separation areas where it was dissolved using acids. In the PGM separation areas, the various precious metals were separated from the concentrate mixture and sent to the purification areas to be precipitated and purified. Pt and palladium (Pd) compounds were purified in the precious metals area, while rhodium (Rh), iridium (Ir), ruthenium (Ru) and osmium (Os) compounds were purified in the other precious metals area. The metals were precipitated from the solution in the form of chlorinated salts and removed using filter presses and glove boxes. Next, the chlorinated precious metals salts were ignited in the ignition areas to form a metal sponge and crushed into specific sizes in the crushing areas. The metal sponge or crushed particles were then either melted into bars or packaged for shipping. Other production activities included melting of Pt and packaging of PGMs since both activities involved the postprocessing of Pt metal. Other non-production activities included handling laundry, laboratory work and maintenance.

Following a workplace inspection and prior to the sample collection, workers were divided into work area groups according to the location inside the plant where they worked. The workers in the same work area group all worked in similar areas where they were expected to experience similar levels of exposure based on the refineries’ historical respiratory soluble Pt exposure data. Prior to sample collection, workers were also grouped into either a direct or an indirect exposure group. Workers in the direct exposure group were directly involved in production activities and came into direct contact with Pt compounds (concentrate handling, separation, precious metals, other precious metals and maintenance workers). Workers from the indirect exposure group were not directly involved in process activities but still came in contact with Pt compounds through indirect pathways while performing laundry, security, laboratory and health clinic activities. Laundry workers were exposed to Pt compounds while washing contaminated clothing. Security workers were exposed while inspecting work areas or searching individuals entering or leaving work areas. Laboratory workers handled samples containing Pt compounds and health clinic workers consulted with workers whose clothes were contaminated.

### Collection and analysis of urinary Pt

Schierl *et al*11 reported that the urinary Pt excretion of volunteers who handled ammonium hexachloroplatinate salts reached the maximum approximately 10 hours after exposure. Since this was similar to the present study where workers predominantly handled ammonium hexachloroplatinate salts, spot urine samples were collected in the morning, approximately 16 hours after the cessation of the previous shift’s exposure. This represented the maximum urinary Pt excretion resulting from the previous day’s exposure. Spot urine samples were collected on three consecutive mornings, prior to the start of the shift. No ‘baseline’ or ‘before exposure’ urine samples could be obtained, and the three spot urine samples represented the urinary Pt excretion of workers over a 48-hour period during normal working conditions. Schierl *et al*11 and Weber *et al*10 reported that workers’ urinary Pt excretion did not decrease following 2 weeks of vacation. As a result, and because some of the workers also worked on weekends, urine samples were not collected on the same days (ie, the same day following the weekend) for all workers. Workers washed their hands prior to sample collection. Complete void urine samples were collected in sterile urine containers, after which a representative 20 mL of the sample was decanted into a sterile suitable high-density polyethylene bottle, which was supplied by the analytical laboratory (Ampath Laboratories, South Africa). All urine samples were collected at the health clinics of the refineries and were sealed immediately following collection in order to prevent contamination. Urine samples were frozen following collection and analysed by Ampath Laboratories using inductively coupled plasma-mass spectrometry. The limit of detection (LOD) for Pt in urine was 0.1 µg Pt/L.

### Statistical data analysis

Statistical analysis was carried out using Statistica V.13.2 (StatSoft, Palo Alto, California) and SPSS V.25. Figures were created using GraphPad Prism V.6.0 (GrapPad Software, San Diego, California). Results for measurements that were below the LOD for the analytical method were substituted using β-substitution, a substitution method recommended by Ganser and Hewett.24 Urinary Pt concentrations were not normally distributed and were, therefore, log-transformed prior to statistical analysis. All results were expressed as µg Pt/g creatinine, in order to account for the degree of dilution of the urine samples, by dividing the mass of Pt (µg) by mass of creatinine (g) in the urine sample. However, since the majority of published studies only reported concentrations of urinary Pt excretion in µg Pt/L, concentrations were also expressed in µg Pt/L, in order to facilitate comparison between the present study and published literature. Pearson correlation analyses were used to perform correlations between the raw (µg Pt/L) and the creatinine-corrected Pt concentrations. Repeated measures analyses of variance (ANOVAs) and interclass correlation coefficients (ICCs) were used to compare the urinary Pt concentrations of the three sampling days. The variation in urinary Pt excretion between different groups was compared for statistical significance using either paired Student’s t-tests (sex, race and direct or indirect exposure groups) or analyses of covariance (ANCOVA) (age, year employed and work areas). The ANCOVA with a Tukey post-hoc test compared the urinary Pt excretion of workers in different work areas while including their years of employment at the refineries as covariates. Effect sizes, through partial eta-squared (η2) values, were used to indicate the practical significance that factors such as years of employment and work area had on the urinary excretion of Pt.25 For example a partial η2 value of 0.5 indicated that the specific variable explained 50% of the variation in urinary Pt excretion. Analyses with a P≤0.05 were considered to be statistically significant.

## Results

Figure 1 illustrates the urinary Pt excretions from the first, second and third spot urine samples collected from all of the participants at the two refineries. A geometric mean (GM) urinary Pt excretion of 0.19 µg Pt/g creatinine (95% CI 0.13 to 0.29 µg Pt/g creatinine) was measured for day 1, 0.23 µg Pt/g creatinine (95% CI 0.16 to 0.35 µg Pt/g creatinine) for day 2 and 0.21 µg Pt/g creatinine (95% CI 0.14 to 0.31 µg Pt/g creatinine) for day 3. A repeated measures ANOVA (P=0.262) and an ICC calculation (0.865) did not show statistically significant differences between the urinary Pt excretions on the three consecutive days of sampling. Additionally, no statistical differences were observed between urine samples that were collected on different days of the week (P=0.648). The urinary Pt excretions from all of the refinery workers (n=40) who participated in the study are summarised in table 1, where the workers were divided into groups according to work areas, exposure categories (direct or indirect exposure), sex, race, age groups and years of employment in the refineries. The urinary Pt excretion concentrations in table 1 are expressed in µg Pt/g creatinine as well as µg Pt/L, and since the data were not normally distributed the GM and the GM’s 95% CIs were calculated. Pearson correlations showed a very strong positive correlation (r=0.948; P<0.001) between the raw and the creatinine-corrected Pt concentrations. A GM urinary Pt excretion of 0.21 µg Pt/g creatinine (95% CI 0.17 to 0.26 µg Pt/g creatinine) was measured for the total group of workers and the concentrations ranged from <0.10 to 3.00 µg Pt/g creatinine. For the total group of workers, 27% of the samples were below the LOD, which is within the range (1%–50%) recommended by Ganser and Hewett24 when using the β-substitution method. The urinary Pt excretions from the 10 participants in the control group were all below the LOD of 0.1 µg/L.

![Figure 1](http://oem.bmj.com/https://oem.bmj.com/content/oemed/75/6/436/F1.medium.gif)

[Figure 1](http://oem.bmj.com/content/75/6/436/F1)

Figure 1 
Summary of the urinary platinum (Pt) concentrations (μg Pt/g creatinine) of the first, second and third spot urine samples. The line in the middle of the box indicates the median concentration. The box extends from the 25th to the 75th percentiles, while the upper and lower limits indicate the maximum and minimum values, respectively.

View this table:
[Table 1](http://oem.bmj.com/content/75/6/436/T1)

Table 1 
Concentrations of platinum measured in urine of precious metals refinery workers

The highest GM urinary Pt excretion was observed in the concentrate handling area (0.58 µg Pt/g creatinine (95% CI 0.34 to 0.98 µg Pt/g creatinine)), followed by the precious metals area (0.53 µg Pt/g creatinine (95% CI 0.29 to 0.97 µg Pt/g creatinine)), and the crushing and ignition area (0.47 µg Pt/g creatinine (95% CI 0.27 to 0.83 µg Pt/g creatinine)). The lowest GM urinary Pt concentrations were found in the other precious metals area (0.07 µg Pt/g creatinine (95% CI 0.05 to 0.11 µg Pt/g creatinine)) and in the group of workers performing other production activities (0.07 µg Pt/g creatinine (95% CI 0.05 to 0.09 µg Pt/g creatinine)).

Figure 2 illustrates the urinary Pt excretions as measured in various work areas as well as the areas where workers were directly or indirectly exposed to Pt. Workers who were directly exposed to Pt had statistically significantly higher urinary Pt excretions compared with workers who were indirectly exposed to Pt (P=0.007). The influence that years of employment had on the urinary Pt excretions of workers was also investigated and was included as covariate in the ANCOVA. The area where workers worked had the largest effect on their urinary Pt excretion (P=0.0006; η2=0.567) while their years of employment also had a statistically significant effect on their urinary Pt excretion (P=0.003; η2=0.261). Therefore, 57% of the variation in workers’ urinary Pt excretion was explained by their work area while 26% was due to the years which they were employed at the refinery. The ANCOVA with a Tukey post-hoc test identified statistically significant differences between the urinary Pt excretions of workers from different work areas (table 1 and figure 2(a–g)). Workers were grouped into categories according to their sex, race, age and the number of years which they were employed at the refinery. A statistically significant difference was observed between the urinary Pt excretions of men compared with women (P=0.007), but no statistical differences were observed between race (Africans and Caucasians) (P=0.206) or between the various age groups (P=0.705). However, after including the work areas as covariate, the difference between sexes was no longer statistically significant (P=0.339).

![Figure 2](http://oem.bmj.com/https://oem.bmj.com/content/oemed/75/6/436/F2.medium.gif)

[Figure 2](http://oem.bmj.com/content/75/6/436/F2)

Figure 2 
Summary of urinary platinum (Pt) excretions (μg Pt/g creatinine) of workers in various work areas of the refineries. The line in the middle of the box indicates the median concentration. The box extends from the 25th to the 75th percentiles, while the upper and lower limits indicate the maximum and minimum values, respectively. The number of samples (n) collected from each work area is indicated above each plot. The letters above the graphs indicate statistically significant differences between the urinary Pt excretions from different work areas as determined by analysis of covariance and Tukey post-hoc test. The brackets above the figure indicate the areas where workers were directly or indirectly exposed to Pt. PGM, platinum group metals.

## Discussion

This study aimed to quantify the urinary Pt excretion of South African precious metals refinery workers over three consecutive work days. The results reflect the Pt body burden of workers that resulted from exposure to Pt compounds via all possible exposure routes as experienced by the workers during the normal routines. Additionally, this study aimed to distinguish between the urinary Pt excretions of different groups of workers in order to identify work areas that pose an increased risk to the health of workers. The following discussions, therefore, identify patterns in the urinary Pt excretion of different worker groups (eg, work area groups) using group statistics. Although the urinary Pt excretions of precious metals refinery and automotive catalyst production workers in Europe have been reported,8 11 15 this is the first study to report the urinary Pt excretion of precious metals refinery workers in South Africa, the world’s largest supplier of Pt.22

Schaller *et al*14 determined that urine can be used to estimate elevated Pt intake, and other investigations conducted in automotive catalyst production plants reported that urinary Pt is an efficient biomarker for occupational biological monitoring and that urinary Pt can be used as a reliable marker for short-term occupational exposure.7 8 The use of urinary Pt excretion in occupational monitoring usually involves the collection of spot urine samples,7 8 11 14 15 which is advantageous, since spot urine samples are non-invasive and easy to collect.26 However, spot samples only provide a brief snapshot of the concentration of a chemical within an individual at a specific point in time and do not necessarily represent the internal concentration over longer time periods. This may lead to misinterpretation or misuse of results, which can lead to the misclassification of the degree of exposure experienced by workers.26–28 According to Smolders *et al*,28 the elimination half-life of the chemical in question, the pattern and intensity of exposure, and the sampling parameters are factors that should be understood when collecting spot urine samples since they can influence the representativeness of the results. Especially the relationship between the chemical’s elimination half-life and the exposure interval is of particular importance.27 The variation in the representativeness of the spot urine samples is the lowest when the chemical’s elimination half-life is long and the exposure to the chemical is frequent, and is the greatest when the chemical’s elimination half-life is short and the exposure to the chemical is infrequent.27 28 The elimination half-life of Pt is long11 and refinery workers are exposed to Pt compounds on a daily basis. Figure 1, along with repeated measures ANOVA and ICC calculations, showed that there was no significant difference between the three sampling days and that the spot urine samples collected during the study did not have a high degree of variability. Additionally, the specific days on which samples were collected also did not influence the results. This corresponds to Schierl *et al*,29 who reported very small intraindividual variability of urinary Pt concentrations during their study and revealed that single spot urine samples could be used instead of 24-hour samples to investigate the urinary Pt excretion of patients who were treated with cisplatin during chemotherapy. Since the urinary Pt excretion of workers did not differ significantly between sampling days, spot urine samples could be used to determine which workers had an increased urinary Pt excretion or Pt body burden, compared with others. Since the elimination half-life of Pt is long,11 it was not practically possible to establish personal baseline urinary Pt excretion values. The results reported in this study, therefore, represent the urinary Pt excretion of 40 workers over a period of 48 hours during normal working conditions.

A creatinine correction was used to adjust for the dilution of the spot urine samples. In a study on the variability of urinary metal biomarkers, Smolders *et al*28 reported that the correction using creatinine or relative density correlated better with the calculated excretion rate of the metals than the raw uncorrected data. However, no previous studies have reported urinary Pt excretion corrected using the relative density of urine. Therefore, the creatinine adjustment was preferred to allow for the comparison with published literature.

Previous studies have associated increased urinary Pt excretion with work areas where the airborne Pt was known to be high.8 11 15 It was observed during this study (figure 2) that the area of work had the greatest influence on the urinary Pt excretions of workers and that the mean urinary Pt excretions of workers who were directly exposed to Pt during production activities were significantly higher than that of workers who came into indirect contact with Pt compounds during non-production activities. Figure 2 also shows that the urinary Pt excretion measured in the concentrate handling area, where PGM-containing material was received, sieved and concentrated, was significantly higher than the urinary Pt excretion measured in areas where workers handled other PGMs (Rh, Ir, Ru and Os) (P=0.021), areas where workers performed other production activities (laboratory, melting and packaging) (P=0.020), as well as areas where workers performed other non-production activities (laundry, laboratory and maintenance) (P=0.032). The urinary Pt excretion measured in the precious metals area, where Pt and Pd salts were handled, was also significantly higher than the results from areas where workers worked with other precious metals (P=0.027) and where workers performed other production (P=0.026) and non-production (P=0.043) activities. The analytical method used during this study (detection limit 0.1 µg Pt/L) allowed for the complete analysis of the urinary Pt excretion of workers who experienced low to high exposure to Pt compounds in various work areas. However, this method was not as sensitive as methods used by other published studies that investigated urinary Pt excretion due to environmental exposure.20 Therefore, the method limited the further analysis of the urinary Pt excretion of workers who experienced very low or environmental exposure (control group) to Pt compounds.

A significant difference was observed between the urinary Pt excretions of men compared with women. However, only two of the eight women worked in production areas (other precious metals area), with the other six performing in non-production activities (indirect exposure). Subsequently these significant differences became non-significant with the inclusion of work area as a covariate. This suggests that the difference between the urinary Pt excretions of men and women was caused by the differences in work area rather than sex-related differences.

The urinary Pt excretion of the refinery workers was influenced by their work areas as well as the number of years which they were employed at the refineries. This is based on the time that workers were employed at the refineries and not the time they worked in specific areas. Schierl *et al*11 measured elevated urinary Pt excretions in workers 2 to 6 years after their previous exposure episodes and suggested that, following exposure, a reservoir of Pt could form in the body similar to that found in patients following treatment with cisplatin.29 Pt has been reported to accumulate in various organs in the body. Following the exposure of rats to Pt compounds via inhalation, the highest concentrations of Pt were found in the lungs and the kidneys.30 The analysis of the Pt content of non-occupationally exposed human autopsied tissue found the highest concentrations in the subcutaneous fat, the kidneys, the pancreas and the liver.31 In vitro permeation studies reported that most of the Pt was retained inside the skin, leading the authors to state that a Pt reservoir may form inside the skin from where Pt could be gradually released. Therefore, following occupational exposure, Pt can accumulate in various organs in the body. It might be possible that working in high exposure areas within the refinery for many years increases the Pt reservoir in the various organs in the body, which could then be gradually released into the systemic circulation and lead to increased urinary Pt excretion. However, it is unknown how the accumulation of Pt at various sites (ie, lungs, kidneys, skin, subcutaneous fat and others) will affect the excretion pattern of Pt from the body.

The urinary Pt excretions observed during this study (<0.10 to 3.00 µg Pt/g creatinine or <0.10 to 5.40 µg Pt/L) are comparable with the results reported by other studies conducted in precious metals refineries and automotive catalyst production plants. Johnson *et al*32 and Farago *et al*15 reported urinary Pt excretions for precious metals refinery workers of between <0.1 and 2.58 µg Pt/L and between 0.21 and 1.18 µg Pt/g creatinine, respectively. Schierl *et al*11 reported urinary Pt excretions of between 0.016 and 6.27 µg Pt/g creatinine in a refinery and catalyst production company where spot urine samples were collected at the end of the shift and on the morning following exposure. Weber *et al*10 and Schaller *et al*14 reported urinary Pt excretions for automotive catalyst production workers of between 0.01 and 2.90 µg Pt/L and between 0.02 and 9.20 µg Pt/L, respectively. More recently, Cristaudo *et al*8 reported mean urinary Pt excretions in an automotive catalyst production plant ranging from 0.10 µg Pt/L in the administrative area to 1.86 µg Pt/L in the coating area.

## Conclusion

This was the first study to report quantifiable concentrations of Pt in the urine of South African precious metals refinery workers. The urinary Pt excretion of workers did not vary statistically between the different sampling days. This was consistent with other metal biological monitoring studies which stated that a long elimination half-life of a particular metal and frequent exposure to it will cause its concentration in spot urine samples to have a low degree of variability.

Similar to previous studies, this study reported significant differences between the urinary Pt excretion of workers from various work areas within the refineries. Significantly higher urinary Pt excretion was observed in areas where workers were directly exposed to Pt compounds during production activities compared with areas where they were indirectly exposed during non-production activities. The urinary Pt excretion of workers can, therefore, be used to differentiate between workers who are directly or indirectly exposed to Pt or workers who work in high or low exposure areas. Additionally, the number of years of employment at the refineries also influenced their urinary Pt excretion. However, this association needs to be interpreted with care since it does not refer to the time worked at specific areas but the time worked at the refinery.

During this study, workers who directly handled soluble Pt compounds and who worked in areas such as the concentrate handling area had increased urinary Pt excretion, which indicated that they experienced increased Pt intake. Since soluble Pt sensitisation has only been associated with increased exposure to soluble Pt compounds, these workers have an increased risk of becoming sensitised.

The urinary Pt excretion of South African precious metals refinery workers observed during this study was comparable with seven other studies conducted in precious metals refineries and automotive catalyst production plants in Europe.

## Acknowledgments

We would like to thank the participants and management teams of the precious metals refineries for their support.

## Footnotes

*   Contributors All the authors participated in designing the study and writing the manuscript. SJLL conducted the survey and the statistical analysis. All the authors are in agreement with the content of the manuscript and with the order of authorship.

*   Funding This work is based on the research financially supported by the National Research Foundation of South Africa (NRF) grant numbers 90562 and 105636.

*   Disclaimer Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors, and therefore the NRF does not accept any liability in regard thereto.

*   Competing interests None declared.

*   Patient consent Obtained.

*   Ethics approval Ethics approval for the study was obtained from the Health Research Ethics Committee of the North-West University (NWU-00128-14-A1).

*   Provenance and peer review Not commissioned; externally peer reviewed.

## References

1.  Hunter D, Milton R, Perry KMA. Asthma caused by the complex salts of platinum. Occup Environ Med 1945;2:92–8. [doi:10.1136/oem.2.2.92](http://dx.doi.org/10.1136/oem.2.2.92)
    
    [FREE Full Text](http://oem.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czo1OiJvZW1lZCI7czo1OiJyZXNpZCI7czo2OiIyLzIvOTIiO3M6NDoiYXRvbSI7czoyMDoiL29lbWVkLzc1LzYvNDM2LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

2.  Baker DB, Gann PH, Brooks SM, et al. Cross-sectional study of platinum salts sensitization among precious metals refinery workers. Am J Ind Med 1990;18:653–64. [doi:10.1002/ajim.4700180604](http://dx.doi.org/10.1002/ajim.4700180604)
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=2264564&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 
    
    [Web of Science](http://oem.bmj.com/lookup/external-ref?access_num=A1990EH87300003&link_type=ISI) 

3.  Calverley AE, Rees D, Dowdeswell RJ, et al. Platinum salt sensitivity in refinery workers: incidence and effects of smoking and exposure. Occup Environ Med 1995;52:661–6. [doi:10.1136/oem.52.10.661](http://dx.doi.org/10.1136/oem.52.10.661)
    
    [Abstract/FREE Full Text](http://oem.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NToib2VtZWQiO3M6NToicmVzaWQiO3M6OToiNTIvMTAvNjYxIjtzOjQ6ImF0b20iO3M6MjA6Ii9vZW1lZC83NS82LzQzNi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

4.  1.  Nordberg GF, 
    2.  Nordberg M
    
    Kiilunen M, Aitio A, Platinum ST, et al. In: Nordberg GF, Nordberg M, eds. Handbook of toxicology of metals. Volume II: specific metals. 4th ed. London, Waltham, San Diego: Academic Press, Elsevier, 2015:1113–23.
    
    

5.  Heederik D, Jacobs J, Samadi S, et al. Exposure-response analyses for platinum salt-exposed workers and sensitization: a retrospective cohort study among newly exposed workers using routinely collected surveillance data. J Allergy Clin Immunol 2016;137:922–9. [doi:10.1016/j.jaci.2015.07.030](http://dx.doi.org/10.1016/j.jaci.2015.07.030)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1016/j.jaci.2015.07.030&link_type=DOI) 

6.  Merget R, Kulzer R, Dierkes-Globisch A, et al. Exposure-effect relationship of platinum salt allergy in a catalyst production plant: conclusions from a 5-year prospective cohort study. J Allergy Clin Immunol 2000;105:364–70. [doi:10.1016/S0091-6749(00)90089-7](http://dx.doi.org/10.1016/S0091-6749(00)90089-7)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1016/S0091-6749(00)90089-7&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=10669860&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 
    
    [Web of Science](http://oem.bmj.com/lookup/external-ref?access_num=000085402200024&link_type=ISI) 

7.  Petrucci F, Violante N, Senofonte O, et al. Biomonitoring of a worker population exposed to platinum dust in a catalyst production plant. Occup Environ Med 2005;62:27–33. [doi:10.1136/oem.2004.016022](http://dx.doi.org/10.1136/oem.2004.016022)
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=15613605&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

8.  Cristaudo A, Picardo M, Petrucci F, et al. Clinical and allergological biomonitoring of occupational hypersensitivity to platinum group elements. Anal Lett 2007;40:3343–59. [doi:10.1080/00032710701689099](http://dx.doi.org/10.1080/00032710701689099)
    
    

9.  Aitio A. Guidance values for the biomonitoring of occupational exposure. State of the art. Med Lav 2006;97:324–31.
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=17017366&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

10. Weber A, Schaller KH, Angerer J, et al. Objektivierung und quantifizierung einer beruflichen platinbelastung beim umgang mit platinhaltigen katalysatoren. Verhandlungen der deutschen gesellschaft für arbeitsmedizin. 13 jahrestagung in Berlin. Berlin: Gentner Verleg, 1991:611–4.
    
    

11. Schierl R, Fries HG, van de Weyer C, et al. Urinary excretion of platinum from platinum industry workers. Occup Environ Med 1998;55:138–40. [doi:10.1136/oem.55.2.138](http://dx.doi.org/10.1136/oem.55.2.138)
    
    [Abstract/FREE Full Text](http://oem.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NToib2VtZWQiO3M6NToicmVzaWQiO3M6ODoiNTUvMi8xMzgiO3M6NDoiYXRvbSI7czoyMDoiL29lbWVkLzc1LzYvNDM2LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

12. Franken A, Eloff FC, du Plessis J, et al. *In vitro* permeation of platinum through African and Caucasian skin. Toxicol Lett 2015;232:566–72. [doi:10.1016/j.toxlet.2014.12.010](http://dx.doi.org/10.1016/j.toxlet.2014.12.010)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1016/j.toxlet.2014.12.010&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=25522659&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

13. Franken A, Eloff FC, Du Plessis J, et al. *In vitro* permeation of platinum and rhodium through Caucasian skin. Toxicol In Vitro 2014;28:1396–401. [doi:10.1016/j.tiv.2014.07.007](http://dx.doi.org/10.1016/j.tiv.2014.07.007)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1016/j.tiv.2014.07.007&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=25084315&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

14. Schaller KH, Angerer J, Alt F, et al. The determination of platinum in blood and urine as a tool for the biological monitoring of internal exposure. Proceedings of the international conference on monitoring of toxic chemicals and biomarkers. Berlin, Germany: SPIE, 1992;1716:498–504.
    
    

15. Farago ME, Kavanagh P, Blanks R, et al. Platinum concentrations in urban road dust and soil, and in blood and urine in the United Kingdom. Analyst 1998;123:451–4. [doi:10.1039/a705920e](http://dx.doi.org/10.1039/a705920e)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1039/a705920e&link_type=DOI) 
    
    [GeoRef](http://oem.bmj.com/lookup/external-ref?access_num=2014098940&link_type=GEOREF) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=9659707&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

16. Petrucci F, Violante N, Senofonte O, et al. Development of an analytical method for monitoring worker populations exposed to platinum-group elements. Microchemical Journal 2004;76:131–40. [doi:10.1016/j.microc.2003.11.005](http://dx.doi.org/10.1016/j.microc.2003.11.005)
    
    

17. Ensslin AS, Huber R, Pethran A, et al. Biological monitoring of hospital pharmacy personnel occupationally exposed to cytostatic drugs: urinary excretion and cytogenetics studies. Int Arch Occup Environ Health 1997;70:205–8. [doi:10.1007/s004200050208](http://dx.doi.org/10.1007/s004200050208)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1007/s004200050208&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=9298404&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 
    
    [Web of Science](http://oem.bmj.com/lookup/external-ref?access_num=A1997XU12700010&link_type=ISI) 

18. Nygren O, Lundgren C. Determination of platinum in workroom air and in blood and urine from nursing staff attending patients receiving cisplatin chemotherapy. Int Arch Occup Environ Health 1997;70:209–14. [doi:10.1007/s004200050209](http://dx.doi.org/10.1007/s004200050209)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1007/s004200050209&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=9298405&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 
    
    [Web of Science](http://oem.bmj.com/lookup/external-ref?access_num=A1997XU12700011&link_type=ISI) 

19. Begerow J, Sensen U, Wiesmüller GA, et al. Internal platinum, palladium, and gold exposure in environmentally and occupationally exposed persons. Zentralbl Hyg Umweltmed 1999;202:411–24.
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=10546331&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

20. Iavicoli I, Bocca B, Carelli G, et al. Biomonitoring of tram drivers exposed to airborne platinum, rhodium and palladium. Int Arch Occup Environ Health 2007;81:109–14. [doi:10.1007/s00420-007-0195-y](http://dx.doi.org/10.1007/s00420-007-0195-y)
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=17492463&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

21. Kopp B, Crauste-Manciet S, Guibert A, et al. Environmental and biological monitoring of platinum-containing drugs in two hospital pharmacies using positive air pressure isolators. Ann Occup Hyg 2013;57:374–83. [doi:10.1093/annhyg/mes073](http://dx.doi.org/10.1093/annhyg/mes073)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1093/annhyg/mes073&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=23091112&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

22. Matthey J. PGM Market Report November 2016. 2016 [http://www.Pt.matthey.com/documents/new-item/pgm\_market\_reports/pgm\_market\_report\_november\_2016.pdf](http://www.Pt.matthey.com/documents/new-item/pgm\_market\_reports/pgm\_market_report_november_2016.pdf) (accessed 4 Apr 2017).
    
    

23. Anand SS. Using ethnicity as a classification variable in health research: perpetuating the myth of biological determinism, serving socio-political agendas, or making valuable contributions to medical sciences? Ethn Health 1999;4:241–4. [doi:10.1080/13557859998029](http://dx.doi.org/10.1080/13557859998029)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1080/13557859998029&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=10705561&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

24. Ganser GH, Hewett P. An accurate substitution method for analyzing censored data. J Occup Environ Hyg 2010;7:233–44. [doi:10.1080/15459621003609713](http://dx.doi.org/10.1080/15459621003609713)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1080/15459621003609713&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=20169489&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

25. Ellis SM, Steyn HS. Practical significance (effect sizes) versus or in combination with statistical significance (p-values). Management Dynamics 2003;12:51–3.
    
    

26. Wang YX, Feng W, Zeng Q, et al. Variability of metal levels in spot, first morning, and 24-hour urine samples over a 3-month period in health adult Chinese men. Environ Health Perspect 2014;124:468–76.
    
    

27. Aylward LL, Hays SM, Smolders R, et al. Sources of variability in biomarker concentrations. J Toxicol Environ Health B Crit Rev 2014;17:45–61. [doi:10.1080/10937404.2013.864250](http://dx.doi.org/10.1080/10937404.2013.864250)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1080/10937404.2013.864250&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=24597909&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

28. Smolders R, Koch HM, Moos RK, et al. Inter- and intra-individual variation in urinary biomarker concentrations over a 6-day sampling period. Part 1: metals. Toxicol Lett 2014;231:249–60. [doi:10.1016/j.toxlet.2014.08.014](http://dx.doi.org/10.1016/j.toxlet.2014.08.014)
    
    

29. Schierl R, Rohrer B, Hohnloser J. Long-term platinum excretion in patients treated with cisplatin. Cancer Chemother Pharmacol 1995;36:75–8. [doi:10.1007/BF00685736](http://dx.doi.org/10.1007/BF00685736)
    
    [CrossRef](http://oem.bmj.com/lookup/external-ref?access_num=10.1007/BF00685736&link_type=DOI) 
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=7720180&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 
    
    [Web of Science](http://oem.bmj.com/lookup/external-ref?access_num=A1995QR54700013&link_type=ISI) 

30. Moore W, Malanchuk M, Crocker W, et al. Whole body retention in rats of different 191Pt compounds following inhalation exposure. Environ Health Perspect 1975;12:35–9. [doi:10.1289/ehp.751235](http://dx.doi.org/10.1289/ehp.751235)
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=1227859&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

31. Vandiver F, Duffield FV, Yoakum A, Bumgarner J, et al. Determination of human body burden baseline date of platinum through autopsy tissue analysis. Environ Health Perspect 1976;15:131–4. [doi:10.2307/3428395](http://dx.doi.org/10.2307/3428395)
    
    [PubMed](http://oem.bmj.com/lookup/external-ref?access_num=1001291&link_type=MED&atom=%2Foemed%2F75%2F6%2F436.atom) 

32. Johnson D, Prevost R, Tillery J, et al. Baseline levels of platinum and palladium in human tissue. San Antonio Texas: Southwest Research Institute, 1976.