A study to characterise the “arsenic rash” observed at a copper smelter in Tsumeb, Namibia

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
This study is located at a copper smelter. Arsenic is a component of copper bearing ore and arsenic trioxide is a by-product of copper smelting. The vapours (“off-gases”) that are released from the molten copper-bearing ore cool and condense in evaporative coolers to form arsenic-containing dust in the smelter's flues and stacks. The dust is filtered in “bag houses” and the captured powder is transported to Godfrey Roasters where the arsenic trioxide is driven off by heat. From there, the hot roaster gases are collected in what are known as “arsenic kitchens”, where fume is allowed to cool. The arsenic trioxide settles to the floor of the rooms as coarse dust and also forms crystalline deposits on the walls and ceilings. It is removed from the kitchens and prepared for shipment as a dust containing approximately 98% arsenic trioxide. Workers are exposed to arsenic containing dust during maintenance work on the copper smelter's flues and the bag houses, or whilst performing various tasks in the arsenic plant roasters and kitchens. The smelter has approximately 500 permanent employees and variable numbers of contractors, which can increase the total employee compliment by 1000. Skin rashes are the most common occupational disease reported at the copper smelter. The underlying cause of these acute transient rashes have historically been attributed by smelter employees to arsenic exposure, as encapsulated in the widely-used term, “arsenic rash”. Previous smelter reports have shown that the highest rates of skin rashes occurred at the arsenic plant and ausmelt baghouse. The appearance and anatomical distribution of the rash was described in these reports. Notwithstanding the use of the term, “arsenic rash”, the role of arsenic trioxide (As2O3) in the development of these rashes has been uncertain. In particular, a question has been raised as to whether the rash represents an allergy to either the arsenic trioxide or some other constituent of baghouse dust. Other uncertainties have related to the roles of skin hygiene practices and Personal Protective Equipment (PPE). This study included three levels of enquiry: a detailed questionnaire (exploring personal risk factors, skin hygiene practices, PPE use and descriptions of the rashes experienced); a clinical examination by a dermatologist; and skin patch testing (using both standard allergens as well as selected chemical agents from 4 selected workplaces, namely “pure” As2O3 powder from the arsenic plant, baghouse dust from the ausmelt & convertor plants, and “cake” from the Effluent Treatment Plant (“ETP”)). Analysis of the chemical compounds present in the four samples was performed by an external certified laboratory in Pretoria, South Africa. For the purposes of skin patch testing, all four samples were “standardised” to 2g/dL As2O3 in water by the Smelter's quality assurance lab. The epidemiological techniques varied according to the different objectives. For objectives 1 and 2: retrospective case control. For objective 3: exposure characterisation for use in exposure-response analysis. For objectives 4 & 5: retrospective case-control study, with controls matched for area of work in the smelter. Cases (N=27) comprised all employees who had one or more work-related rash incident within the preceding 12-24 months. These are relatively rare events, limiting the number of cases available for the study. Controls (N=24) comprised purposively selected co-workers, one control for each case, who performed the same type of work in the same workplace but who had never developed a rash. The principal variables included potential determinants of skin reactions to workplace materials (pure arsenic dust from the arsenic plant, baghouse dust from the ausmelt and converter and filter cake from the effluent treatment plant), reactions to patch testing, a history of allergy, prior experience of a similar rash, duration of service in the smelter and age. Ethical approval to conduct the study was obtained from the University of Cape Town's Health Research Ethics Committee (UCT HREC) (reference number 261/2016), and from the Office of the Permanent Secretary for Health, Namibia (letter dated 17 January 2018). Five study objectives were formulated, the outcomes of which are summarised as follows: Study objective 1: To interrogate if skin hygiene and hand cleansing practices used in the smelter, notably the use of barrier creams and soaps, are risk factors for developing a skin rash, and if so, at which anatomical location. The study data showed that hand washing practices of cases and controls were very similar, suggesting that handwashing practices are not a risk factor for developing a skin rash at any anatomical location, notably the hands. Furthermore, the data showed that the hand cream being issued as a barrier to chemical contact is not protective. However, both findings could be due to a non-differential bias whereby responses to the questionnaires in both groups of participants were influenced by a desire to appear compliant with company policy and conscientious with regard to cleanliness. Study objective 2: To interrogate the use of PPE by employees in the smelter, and whether or not this is a contributory factor to the development of the rash Whilst responses for the individual PPE related questions were generally similar in cases & controls, the combined prevalence of rashes in the area of the face (41%), respirator contact points (12.5%) and the neck area (19.6%) is high (73%). The rest of the body combined only accounts for 27%. Also, more specific questioning of the cases suggests that the respirators are a substantial contributor to the rashes in the face & neck areas. The lack of statistical significance between cases and controls for the individual PPE related questions could be due to the same non-differential bias operative in objective 1. Study objective 3(a & b): To characterise the nature of the chemical constituents in the production byproducts obtained from the various workplaces of the smelter operations (“workplace materials”). These “workplace materials” are the substances (usually in dust form) to which employees are exposed and which may trigger the skin reactions. The analysis addressed this in two ways; objective 3a looked at the chemical constituents of the workplace materials “as-is” (taken from the samples collected directly from the various workplaces as part of the smelter's Quality Assurance (QA) programme, and therefore as they are experienced by workers), and objective 3b looked at the chemical solutions used in the study skin tests, after standardisation for arsenic trioxide at 2g of arsenic trioxide per 100mL of water. Objective 3a: The proportionate concentrations of As2O3 varied from 5% to 98% across the 4 samples from the 4 workplace locations, namely the arsenic plant (98%), ausmelt baghouse (83%), convertor baghouse (21%) and ETP (5%). Lead and sulphur were identified as additional potential irritants in the convertor baghouse dust and the ETP cake. Both baghouse dusts (ausmelt & converter) had alkaline pH, the As2O3 sample from the arsenic plant had an acid pH and the pH of the ETP sample was close to neutral. Objective 3b: The samples were standardised to 2g/dL As2O3 in water, to better ascertain the skin responses to arsenic trioxide specifically at varying dilutions. This concentration was chosen because it is the point of solubility of arsenic trioxide in water. Consequently, the two samples with relatively lower As2O3 in the source material (converter Baghouse dust & ETP cake) had proportionately increased concentrations of their non-arsenic constituents after standardisation. These proportionate increases were 19% (ausmelt baghouse), 373% (converter baghouse) and 1799% (ETP). Should any of the non-arsenic constituents be irritants, their irritancies would be equivalently affected. Following standardisation for arsenic, the pH for the ausmelt sample went up from pH 7.8 to pH 8.8, the converter sample went from pH 8.7 to pH 9.6 and the ETP sample went from pH 6.7 to 7.8. The arsenic plant sample had a pH of 4.5 after standardisation. Unfortunately, the smelter lab did not provide a pH of the pre-standardised arsenic plant sample. The high pH levels (8.8 & 9.6) or low pH level (4.5) are independently capable of causing irritation. Study objective 4: To characterise the nature of the dermatological response to these exposures, notably whether the reactions are allergic or irritant in nature. The main finding of this study is that arsenic trioxide is an irritant not an allergen, because of its low pH as well as an inherent dermal toxicity. The grounds for this conclusion are (1) the clinical appearance of the skin reactions where arsenic trioxide was in contact with the skin and (2) the dose-response relationship with increased concentrations of arsenic trioxide with the skin. The presence of arsenic trioxide in the baghouse dust and ETP cake confers irritant properties to these operational materials. The baghouse dusts are additionally irritant because of their high pH. The ETP cake produced a dosedependent irritant reaction even though it was pH neutral. Irritancy has implications on exposure prevention, in that all employees are potentially affected, not only a subset of vulnerable people. Study objective 5: To try to ascertain any causal relationship between baghouse dust (and notably the As2O3 in the dust) and the pathological outcomes The irritancy of arsenic trioxide in the arsenic plant sample was clearly demonstrable even though it was significantly diluted during the standardisation process (to 2g/dL). Arsenic plant workers in the realworld setting are exposed to undiluted concentrations of arsenic trioxide dust, which explains the high prevalence of irritant skin reactions amongst workers in this area. This study has demonstrated that the alkaline pH of baghouse dust confers additional irritancy to that already conferred by the arsenic trioxide present in the dust. This explains the high prevalence of irritant skin reactions amongst workers exposed to baghouse dust. This report ends with some recommendations, based on the findings of the study, as well as knowledge gaps identified.