|
Survey and Health Assessment of Possible Health Hazardous Compounds in Proofing Sprays 2 Literature retrieval and information search
2.1 IntroductionThe first phase of the project comprised the collection of literature data with regard to known registered information about toxification of consumers during the use of proofing spray. In addition to human cases of toxification this report also includes a number of studies on animals during the use of as a rule commercial proofing sprays with a more or less well-known composition. There are two reasons for that: 1) to procure more information about ingredients in spraying products that have been involved in human cases of toxification, and 2) to get the opportunity to explain certain effects through pathological investigations. A wide range of different proofing agents exist and they are sold directly to the consumers as agents for restorative treatment of different types of textiles, most often to obtain water and stain repellency. The main part of the products is sold as sprays. Cases of poisoning have often been reported in connection with the use of these types of products. In the newest, larger case from Denmark, 10 people became ill within two months in 2005 as a result of using a certain product. The first phase of the project has the following objective:
2.2 Preliminary searchesIn order to create an outline of which substances and/or substance groups it would be relevant to target the payment database literature survey against, a number of preliminary internet searches on relevant Danish and foreign homepages were carried out. The internet searches were carried out by means of the search machine Google and in one particular case it was subsequently chosen to use the same search word in Google Scholar which focuses on scientific references. 2.2.1 Google searchesA number of Danish cases of toxification have been registered in connection with using different types of proofing sprays and therefore a Danish keyword was initially used for the searches. For this preliminary screening the following Danish words were used: spray forgiftning imprægnering, as the word tekstilspray combined with forgiftning did not give any search results. That only resulted in few interesting results. The English search combination was: fluor resin textile spray pulmonary poison, and the advanced search strategy - that all words had to be found - was employed. That combined with results in English gave a total of 14.100 results. 2.2.1.1 Selected search results in GoogleThe Danish search localised a scientific article from the Danish magazine Ugeskrift for Læger from 1999 (Jacobsen et al., 1999). That reference contains a summary of the chemical composition of a number of proofing products. The examination of the compositions was instituted by Giftinformationen (Danish Poison Information Centre at Bispebjerg Hospital: Clinic for Occupational and Environmental Medicine) and it was carried out by the Danish Emergency Management Agency (Beredskabsstyrelsen). It is appropriate to divide the substances found in the investigated products into 3 main groups:
The propellants consist of low molecular hydrocarbons such as propane, butane and isobutane. Earlier, the so-called CFC gases (fluorine and chlorine containing hydrocarbons) were used. The proofing agents can be siloxane compounds, fluorcarbons, urethanes, esters/wax or phthalates. The solvents are typically mixtures of aliphatic hydrocarbons (e.g. heptane isomers) and cyclic hydrocarbons (e.g. cyclohexane) and chlorinated hydrocarbons (e.g. 1,1,1-trichloroethane) and esters (e.g. butylated acetate). Butylated acetate is forbidden in products for indoor household use and 1,1,1-trichloroethane is forbidden in spray cans as it is ozone layer decomposing. Therefore, these two solvents are forbidden in spray cans intended for indoor household use. The searches in English gave two usable results. One Japanese article (Jinn et al., 1998) reports the following content of spray cans: 1,1,1-trichloroethane, liquid petroleum gas (low molecular alkanes) and fluorine based polymers (fluoride resin). In this case, it is a proofing spray that has caused lung injury. The other article (Lazor-Blanchet et al., 2004) does not mention a textile spray but an agent to treat floors (tiles) so discoloration is prevented. The proofing substance in this agent is stated to be: < 1 % acrylic ester fluorpolymers dissolved in a > 90 % mixture of isoalkanes (C9-C12). This product does not exist in a spray can with propellant but is intended for application with a brush. However, the professional tiling company that was mentioned in the article had chosen to fill the liquid into a container with pump spray and apply the agent in that way resulting in toxification. The same acrylate fluorpolymer as in this product (and from the same manufacturer) is in article[3] stated to have caused a number of respiratory problems in connection with household proofing of leather and textiles. 2.3 Bibliographical database searchesIn order to involve knowledge about international experience, a goal-oriented search was carried out in a suited cluster of literature databases (TOXCENTER) and at the same time in a couple of the large databases EMBASE and SCISEARCH that do not form part of this cluster at the database host STN (see description under 2.3.1.2). The search was carried out on the combination of textile proofing and/or the identified chemical substances in relation to the registered symptoms including the word toxification. 2.3.1 Goal-oriented literature searches2.3.1.1 Preparation of search profileOn the basis of keywords from Vernez et al. (2006) and literature references in that article, it was chosen to search for cases of toxification where the following words and word combinations appeared:
The search terms for these parameters are: Acute Respiratory Syndrome OR Lung Injury OR Pulmonary Toxicity OR Pulmonary collapse OR Pneumonia OR Respiratory disease. The mentioned cases of toxification can occur by exposure to the following:
as it was established that the term "textile" had a limiting effect on the number of search results: Other conditions that might manifest themselves are:
2.3.1.2 Payment database searchesThe above search profile was used to search in the below databases. TOXCENTER (Toxicology Center) is a bibliographical database that covers the pharmacological, biochemical, physiological and toxicological effects from medicines and drugs and other chemicals. EMBASE (Excerpta Medica) is a bibliographical database covering literature within the biomedical and pharmaceutical field. Science Citation Index (SciSearch®) contains all recordings published in Science Citation Index ExpandedTM. The search resulted in 9 references, see chapter 8. 2.3.1.3 No cost bibliographical database searchesThe complete search term: ((Acute AND Respiratory AND Syndrome) OR (Lung AND Injury) OR (Pulmonary AND Toxicity) OR (Pulmonary AND collapse) OR Pneumonia OR (Respiratory AND disease)) AND ((Proofing AND Spray) OR (Waterproofing AND spray) OR (Spray AND Impregnation)) was also used in PubMed and on Scirus.com. This very specific search gave 8 search results in PubMed and 11 results in Scirus related to lung effects arising after having used proofing spray. There was a certain overlapping between the references from the payment databases and the no cost databases. The entire reference list can be seen in chapter 8. 2.3.2 Articles and references referred toVernez et al. (2006) and several of the procured articles contain a number of references to additional literature. The complete bibliography of the project comprises a score of references to scientific investigations; see the reference list (chapter 9). However, several of the references relate to proofing products to be used on other materials than textiles. The identified relevant literature was purchased with a couple of exceptions where repeated attempts to place an order gave no result. In addition, three of the identified articles were commented on from the English abstract as the original article was in Japanese. The literature has been investigated in order to identify possible cause and effect relationships between toxifications/symptoms and exposure to chemical substances (isolated substances or combinations) and/or the physical characteristics of aerosols, also including the special conditions for nanoaerosols. Only one of the scientific articles dealt with measurements of the aerosol diameter. In addition, Vernez et al., (2006) have a rather rough measurement of the size distribution on mass basis. Most of the articles discuss the creation of very fine aerosols when spraying liquids from propellant cans and mention that this condition can contribute to the registered lung effects. 2.4 Results2.4.1 Data from referred to articles2.4.1.1 Product compositionThe collected information is presented in Table 2.1. Table 2.1 Complete outline of accessible composition data from procured literature.
Liquid petroleum gas is a mixture of low molecular hydrocarbons – presumably most propane, butane and isobutane. Some of the studies that were found do not contain information about the proofing liquid composition and therefore they have not been included in the table. 2.4.2 Assessment of reported cases of toxificationAll the cases of toxification, reported for proofing spray in the found references, have in common that the products previously have been used without reported lung injuries, often for several years. It is also a common trait that a formulation change of the product has taken place immediately before the observed cases of toxification. That has often taken place with reference to the solvents or propellants being harmful to the environment, and therefore they had to be replaced. The solvents less harmful to the environment and subsequently allowed have often not been able to dissolve a sufficient amount of the originally used water repelling proofing agents and therefore they have been replaced with other substances. That has i.a. been reported in investigations from Switzerland, France, Denmark and the USA (Vermez et al., 2006; BfR, 2006a; Gregersen et al., 2006; Smilkstein et al., 1992; Kulig et al., 1993). On the whole, respiratory injuries connected with the use of proofing spray were observed in a number of cases (Burkardt et al., 1996; Tagawa et al., 2003). Many of the other references describe individual cases (Tanino et al., 1999; Kobayashi et al., 2006). Several of the references stress that tobacco was smoked at the same time as spraying took place or that cigarettes were held between the fingers which still had surplus proofing liquid on them. This is with reference to Teflon compounds (being fluorcarbon polymers) are known to cause ”polymer fume fever” when heated and cases of pulmonary edema owing to pyrolytic products from these polymers (Jinn et al., 1998) have been reported. In two issues of Morbidity and Mortality Weekly Report from 1993, there is reference to poisoning with leather proofing spray in Oregon (Smilkstein et al., 1992) and an "epidemiological note" from Colorado concerning three cases (Kulig et al., 1993). However, in the editorial comment after the actual reports at least 157 cases of consultations to doctors were registered concerning toxification with the same product in the USA. In both cases, the editorial states that a formulation change of the product had taken place shortly before, as the use of 1,1,1-trichloroethane was to be phased-out before 1994 according to the change in the Clean Air Act in 1990. The composition of the leather spray liquids involved in the cases of toxification appears from Table 2.1. The reported composition of spray liquid corresponds to the combination of several textile spray liquids and has therefore been included here. Through questionnaires used as follow-up on a Swiss collection of reported cases of toxification (approx. 200 cases), Vernez et al. (2006) retrospectively investigated to which degree the exposure concentration had influenced various parameters e.g. with regard to consultations at doctors/casualty wards. In the cases where the hospital had been visited the results of the clinical investigations and analyses carried out at the hospital have been further investigated. On the basis of the questionnaires, individual exposure data was generated from a classic 2 zone model for aerosol dispersing in the community and in the distance during use. The resulting evaluated dosage and exposure data were spread over several sizes. A connection was not found between exposure and indicators of health effects (own perception of the seriousness and clinical indicators). A minor connection was found between unspecified inflammation indicators e.g. leucocytes and C reactive protein (a test that measures the blood’s content of a protein that indicates an immediate inflammation) and the maximum exposure concentration. The found results demonstrated that there was considerable individual variation indicating that one or more indirect mechanisms determine the development of the respiratory problems. No threshold value was found for safe exposure. That indicates that increased requirements to the surroundings (ventilation, through draught, room size) during use are not enough to prevent future outbreaks of toxification with proofing spray. The authors conclude that additional precautions have to be taken when marketing new spray products. 2.4.3 Other information from procured literatureYamashita and Tanaka (1995) investigated the administration of aerosols in a number of female mice from the CD-1 strand. They found that prescriptions containing fluor resin caused immediate respiratory disease but none of the other ingredients worked that way. They refer to recent preceding cases of toxification and discuss that changes in solvents ease the creation of aerosols and give a smaller drop size. That could explain the increased toxicity of the reworded spray liquid. A couple of years later, Yamashita et al. (1997) in CD-1 female mice again investigated the toxicity of a spray liquid that had been made water-repellent with fluorcarbon resin. This time different average aerosol diameters in the spray mist were tested. The article demonstrates that the aerosol size is of great importance. When the aerosol diameter increased to 89.1 µm with 0.2 % of particles with a diameter less than or equal to 10 µm, there was no toxicity of the fluorcarbon resin. When the average aerosol diameter in the spray mist was 62.0 µm with 1.6 % of aerosols with a diameter less than or equal to 10 µm there were on the contrary many toxic lung changes in the mice. Tashiro et al., (1997) investigated the effect of a textile spray containing perfluoralkylethylacrylate/n-alkylacrylat copolymer as proofing agent on rats. A sample was taken of the severally damaged surface mucus in the lungs of the rats. Then the group investigated if it was possible to administrate new surface mucus. The objective of the test was to investigate if it was possible to treat damaged lungs through inhalation of an aerosol of lung surface mucus (from pigs). At the same time, the test demonstrates that a commercially available textile spray is very damaging to rat lungs. Hubbs et al. (1997) partly investigated the product composition and partly investigated how a proofing agent (for leather) effects guinea pig and rat lungs. After rewording, the product had been the cause of many respiratory diseases in humans. The previous product caused no toxic changes in guinea pig or rat lungs. The new spray product caused quick breathing, pulmonary edema, haemorrhage from the lung and one death in the exposed guinea pigs and rats. The electro microscopic samples showed direct cytotoxicity in the lungs with alveolar necrosis in type 1 cells and interstitial edemata certain places in the lungs and no effects in other samples. The test demonstrated that the old product with fluoralkenes did not show lung toxicity, but the new product that also contained fluoralkenes demonstrated toxicity in guinea pigs as well as in rats. The change in the composition of the product took place in connection with the phasing-out of 1,1,1-trichloroethane (Clean Air Act amendment from 1990). 2.4.4 NanoaerosolsHere, nanoaerosols are used as the term for small (<100 nm) units of substances or material that are suspended in air and that are not gaseous. Liquid or solid materials can be in question, including amorphous structures. As already mentioned, it was not possible to find information about possible health effects from spray with nanoaerosols – apart from the press release mentioned below. At an expert meeting on 7 April 2006, the German federal agency for Risk Assessment (BfR) discussed if they could find the reason for 97 cases of toxification, of which some were serious, caused by two new sealing spray products that contained nanoparticles (BfR, 2006a). The expert meeting analysed to which extent respiratory problems and pulmonary edema could have been caused by the nanoparticles in the 2 products or if other dangerous substances from traditional proofing agents could be responsible. As the suppliers of the 2 products were unable to supply complete product declarations, it was not possible to carry out a discussion on a sufficiently scientific basis. However, it was agreed that a classic toxiological weighing-out of the individual compound in a mixture is not enough when the product is applied from an aerosol spray with propellant. Here, physical factors such as e.g. drop size play a decisive role for toxic effects in the respiratory passages. The health effects of products from a propellant spray can only be determined with a test strategy that imitates the actual indoor application conditions. Subsequently, (26 May 2006) BfR sent a press release (BfR, 2006b) stating that the two sealing sprays did not contain nanoaerosols (aerosols < 100 nanometer). The reference to "nano" in the marketing of the products was supposed to underline the very thin layer of sealing that was necessary. The cause of the 110 cases of health injuries - of which some were serious - has not yet been established. Therefore, there are for the time being no examples of directly proven toxic lung injuries due to nanoaerosols in spray products. 2.5 Summary of results and conclusionA number of articles were found with information partly about toxic effects in connection with spray proofing and partly about the proofing agents composition with regard to proofing agent, solvent and possible propellant. Some few proofing sprays that caused toxicity in humans were subsequently tested in animals. Many cases of reported toxification from proofing spray have in common that products with the same name previously were used without reported cases of lung injuries. Immediately before the observed cases of toxification a rewording of the product had taken place often in connection with more rigorous environmental laws where the original solvent or propellant was regarded as dangerous to the environment and therefore had to be replaced. The more environmentally friendly solvents that subsequently were used have not been able to dissolve a sufficient amount of the originally used proofing agents which therefore have been replaced by other chemical compounds. In Vernez et al., 2006 there is a rough measurement of the aerosol size distribution on mass basis. In several products that have caused respiratory problems among the users the size of 90% of the aerosol drops was approx. 2-10 µm (Vernex et al, 2006). Most articles that are referred to, discuss why it is important that propellants in aerosols lower the average aerosol diameter in the spray mist. Spray proofing agents involved in reported cases of toxification often contain one or other type of fluorcarbon polymers. There is no actual description of the compounds e.g. in the form of a CAS no. However, in some few American reports there are some chemical describing letter/number references so it should be possible to find detailed descriptions of the chemical structures. On the other hand, solvents and propellants are in general unambiguously described, but amount specifications are rarely in question. Generally speaking and after having gone through the many references, it is still not clear if the registered lung injuries are caused by a kind of immediate chemical pneumonia or if it could be the reaction of the lungs to a fine vaporized hydrophobic mist that penetrates down to the finest bronchioles. Literature lacks data about the aerosol size as well as the chemical composition. When such data has been procured experimental toxicological investigations of the demonstrated substances, including the substances on aerosol basis and the importance of the aerosol size will be necessary. Footnotes[3] Quotation: Interestingly, during the winter 2002-2003, the Swiss Toxicological Information Centre had also recorded an unusual increase in respiratory troubles following household exposure to proofing sprays for conditioning of leather and textiles. After the occurrence of more than 150 such cases, three incriminated aerosols were removed from stores and distribution channels. Investigations by Public Health authorities showed that this outbreak of domestic cases also occurred after a formulation change of the proofing agent. The same new acrylate fluorpolymer produced by the same manufacturer was found as the common component in both our occupational cases and the domestic cases (4). It has not been possible to procure the reference: Office Fédéral de la Santé Publique from 2003.
|