- Email: [email protected]
Manufacturing processes at two French rapeseed oil companies: Possible relationships to toxic oil syndrome in Spain
Fd Chem. Toxic. Vol. 29, No. 12, pp. 797-803, 1991
0278-6915/91 $3.00 + 0.00 Pergamon Press plc
Printed in Great Britain
Research Section MANUFACTURING PROCESSES AT TWO FRENCH RAPESEED OIL COMPANIES: POSSIBLE RELATIONSHIPS TO TOXIC OIL SYNDROME IN SPAIN M. POSADADE L A PAZ*, R. M. PHILENt:~, I. ABAITUABORDA*,J. T. BERNERT,JR:~, J. C. BADAGANCEDO~,P. J. DUCLOS~,[[ and E. M. KILBOURNE:~ *Unidad de Programas de Investigaci6n, Fondo de Investigaci6n Sanitaria, Ministerio de Sanidad y Consumo, Antonio Grilo 10, 28015-Madrid, Spain, ~/Centersfor Disease Control, Public Health Service, US Department of Health and Human Services, Atlanta, GA 30333, USA, §Director del Instituto de Productos, Lficteos de Asturias, Apartado de Correos 85, 33300 Villaviciosa, Asturias, Spain and l[Laboratory Centre for Disease Control, Health and Welfare Canada, Tunney'sPasture, Ottawa, Ontario K1A 0L2, Canada (Accepted 29 July 1991) Abstract--The toxic oil syndrome (TOS) epidemicthat occurred in Spain in spring 1981 has been associated with the consumption of rapeseed oil that was denatured with aniline for industrial use but diverted for human consumption. The precise aetiologic agent in the oil responsible for the outbreak has not been identified. To learn more about possible contaminants and how the contamination might have occurred, we visitedtwo French companies that process rapeseed oil and that were identifiedin Spanish administrative and judicial records as the ones exporting aniline-denatured rapeseed oil to Spain in 1981. With the apparently full and voluntary co-operation of personnel at both companies, we reviewed the processes involvedin manufacturing, treating and transporting rapeseed oil, and we have summarizedthe information provided to us. Of particular importance is the finding that oil exported to Spain was taken from stock, the rest of which was sold for human consumption in the French domestic market, apparently without any adverse health effects. The differences between the oil exported to Spain and the oil sold as food in France were that aniline equivalent to 2% of the weight of the oil was added to most of the Spanish oil but not to that sold in France, and that contamination of the Spanish oil may have occurred in the tank trucks used for transportation to Spain, which had previously carried industrial chemicals. There is no assurance that the trucks were cleaned appropriately for transporting a food product before the oil was loaded for the journey to Spain. Since the clinical manifestations of TOS are not those of aniline toxicity, we conclude that the aetiological agent of TOS is likely to be one of the following: (1) a contaminant in the aniline, (2) a contaminant introduced during transportation, (3) a reaction product of normal oil components or materials used in refiningwith either aniline or the potential contaminants mentioned under (1) or (2) above.
INTRODUCTION
Toxic oil syndrome (TOS) was a previously unrecognized illness that occurred in Spain in epidemic form in spring 1981; it affected more than 20,000 people and caused several hundred deaths (Grandjean and Tarkowski, 1984). Epidemiological investigations have shown that this outbreak was caused by the ingestion of aniline-denatured rapeseed oil initially intended for industrial use, but diverted to human consumption (Cafias and Kilbourne, 1987; Casado Flores et al., 1982; Diaz de Rojas et al., 1987; Kilbourne et al., 1988; Posada et al., 1987; RigauPtrez et al., 1984; Tabuenca, 1981). Although several contaminants have been identified in implicated oils, the precise aetiological agent responsible for TOS
tTo whom all correspondence should be addressed, except reprint requests, which should be addressed to Dr Posada de la Paz. Abbreviations: TPS -- toxic oil syndrome. FCT 29/12--A
has yet to be identified (Bernert et al., 1987; Gardner et al., 1983; Grandjean and Tarkowski, 1984; Hill et al., 1987; Vioque and Vioque, 1983). The transportation and handling of the suspect oil after its importation into Spain has been investigated in detail (Equipo Especial de Investigaci6n, 1981); however, the manufacture and handling of the oil in France has not been described previously. Our review of Spanish administrative and judicial records identified two French companies as the only sources of aniline-denatured rapeseed oil exported to Spain in 1981. We therefore visited these companies to review all steps in the manufacture, treatment and transportation of the implicated oils and to identify any aspects of these procedures that might have led to contamination of oil with the substance or substances that cause TOS. MATERIAI~ AND METHODS
Two of us (M.P. and I.A.) reviewed the six volumes of information that have been collected by Spanish
797
798
M. POSADADE LA PAZ et al.
administrative and judicial authorities regarding the importation of rapeseed oils into Spain in 1981 and previous years. We abstracted data on (1) dates when specific oil shipments left France and entered Spain between January and June 1981; (2) type of oil (whether denatured, the type of denaturant used, and whether refined); (3) quantity of oil; (4) licence numbers of tank trucks used; and (5) other information. We collected data in the same way at the two French companies (companies A and B). At both factories we met with the administrative directors and their associates, as well as with technical personnel directly involved in oil production. We thoroughly inspected the oil production facility and techniques. In particular, we investigated the techniques for adding aniline. We also investigated other possible routes for oil contamination, and typical production runs. The type of information we attempted to elicit is listed below: General activities of the companytypes of activities, volume of production, administrative or technical changes since 1981, products used; seedgenetic variety, geographic origin, content of erucic acid; handling of the seedtransport, quality inspection at arrival, unloading, cleaning, drying, storage; oil processingpressing, solvent extraction, degumming, refining, storage of oil; oil transportcleaning of tankers, transportation companies, final destination(s) of the oils; denaturantscompounds used and chemical characteristics, reasons for changing from one compound to another, governmental regulations governing use of denaturants, mixture of the denaturants with oil, companies involved in the manufacture of denaturants; movement of specific lots of oilnumbers of lots shipped, weights of lots shipped, specific trucks used, shipment dates; Factory operationstypes of machinery present, processing details, temperature, pressure, etc., maintenance of the installations/machinery.
Both companies gave us product samples, including rapeseed oil (crude and refined, if both were available). All samples were of recent origin, since samples from 1981 were not available. RESULTS
In general, the invoices and other records of the two companies agreed well with the data obtained from the Spanish customs authorities. These data related to shipping dates, quantities of denatured oil shipped and the destinations to which the oil was sent. Extracting crude oil from rapeseed is a continuous process with the following steps: (1) cleaning and storing, (2) milling, (3) heating, (4) pressing, and (5) extracting with solvent. The details of these processes are discussed separately for each company. Company A-general
description
On 10 November 1987 we visited company A, which is a subsidiary of a larger holding company that has changed ownership since 198 1. The previous holding company went into bankruptcy for reasons unrelated to the sale of rapeseed oil to Spain. Despite the change, the staff and activities of the company remain the same as in 1981. Company A has three departments that each perform different activities. One department produces and sells crude (unrefined) vegetable oil that is transported in tank trucks; this oil is refined by the buyers and then sold principally for human consumption. The second department produces refined oil for industrial use, and the third department produces resin. Normally, the activities of the three departments are strictly segregated. Only the activities of the first two departments are described in detail here. The factory processes both rapeseed and sunflower seed according to the dictates of the parent (holding) company. It has the capacity to process daily 450 metric tonnes of seed, which produces 180 to 200 metric tonnes of oil a day. According to the technical managers, 98% of the oil is for human consumption, and 2% is for industrial use. The operating facilities and procedures are basically the same as in 1981, when they sold about 500 metric tonnes of rapeseed oil to Spain for industrial use. Company A-seed
The rapeseed is of genetic varieties low in erucic acid and comes to the factory from all over France by truck or rail. Upon arrival, the seed is inspected for quality: the colour and odour is examined, the oil and moisture content is determined, and the oil acidity and erucic acid content is measured. At the factory, rapeseed from different regions and of different varieties are mixed; therefore oil cannot be traced through production records to seed from a particular location or of a specific variety. The plant operators stated that they have no interest in such factors; their concerns are mainly with the oil content and the level of erucic acid. In 1981, the erucic acid content was less than 5%. Although no tests are performed for specific biological or chemical contaminants, no known problems with pesticide contamination have occurred.
799
Rapeseed oil manufacture and TOS Salefur IndustrialUae
Seed f ~ r ~
t
1Ildl~
(~ines)
t
Conk3 1oo*c 15-30 rain
~
t
Sale for Human Consumption
Water Wash and Vacuum Dr)'
t t ~d.
L
Bleached
t
Neutralizationwith Caustic Soda
t
Oepnlmed w~h
t
Ph&pho~c~d
Shipment
Animal Feed Stool Oil (Mixed laddsH)
l
Refined at > CompanyA
Fig. 1. Rapeseed oil production, company A, 1987. After inspection, the seed is unloaded and filtered to separate out stones, pieces of metal, and other foreign material, and is then conveyed to concrete storage silos where it remains for at most 15 days before further processing. There are no facilities for long-term seed storage at the location we visited.
Company A---crude oil production Crude oil is prepared by a standard process, as shown in Fig. 1. Seed is transported from the storage silos to crushing rollers (flaking machines) and then to cookers, which were in operation at about 100°C on the day of our visit. Although sunflower seed was being processed when we were at the factory, we were told that the same conditions (100°C for 15-30 min) are normally used for rapeseed. Cooked seeds are pressed in screw-press expellers, and the expressed oil is then centrifuged, filtered and stored. The press cake from the expellers is conveyed to a separate building where additional oil is extracted with hexane (the only solvent Company A has ever used) by a standard continuous counter-current D'smet solvent-extraction process that has not been altered since 1981. After the hexane is recovered, the extracted oil is stored, often in a mixture with an earlier batch of pressed oil. Only crude oil is prepared by this process; the oil is not degummed or treated in any other way before storage and eventual shipment. The overall crude oil preparation operates continuously, 24 hr a day, 7 days a week, except for *The term used by oil producers to define the concentration of sodium hydroxide used. For liquids heavier than water (such as sodium hydroxide solutions) the Baum~ scale is based on the density of a 10% sodium ehioride solution, which is given a value of 10° Baum~ and the density of pure water which is given a value of 0° Baum6.
a yearly shutdown for general maintenance, usually in July. The crude oil is sold and delivered in tank trucks and trains to refineries that process the oil further before it is sold for human consumption. Before shipment, oil is stored for periods ranging from a few days to 1 month in 1000-tonne capacity tanks. After the solvent extraction process, the remaining press cake is sold as animal feed.
Company A--oil refining At company A, refining oil is a much smaller operation than extracting oil. The oils refined include rapeseed, sunflower, linseed and soybean. Whereas almost all the crude oil produced is intended for human consumption after further refining elsewhere, the refined oil is strictly for industrial use. The refining process used is a common discontinuous (batch) process, as shown in Fig. 1. The steps in refining are as follows: (1) degumming with phosphoric acid, (2) neutralizing with caustic soda (18 ° Baum~*) in four vertical tanks of 15 tonnes, each with gravimetric separation, (3) washing with water and drying under vacuum, and (4) decolourizing (bleaching) with bleaching earth normally obtained from Germany, although an English clay is used occasionally. Spent bleaching earth is removed from the oil by a standard filter press. All bleaching earths used in this process are identical to those used for processing food oils. The industrial oils are not deodorized, hydrogenated, or otherwise modified.
Company A-~lenaturation Tank trucks are inspected and then loaded for shipment to buyers. Trucks that carry consumable oil have inspection cards valid for all of Europe. F o r denatured rapeseed oil, not intended for human
800
M. POSADADE LA PAZ et al. rm,~ x~.,eed ~m, Flum
SeedStorap Co-operatives Companies Steam.Clewed Tanker Trucks Transportby Truck or r~od TmKen
( o.sof
Production]
DENATURED OIL Aniline Added
Storage (Dried and Aerated)
Process 10-TonTankfor Oeanturization
MIIIInl (Flaking) Hexlne
\ Dried Press 4 Cake Stered
HexaneExtraction
t
4 Screw Presses
1
Ip PreooCake
1
UnrefinedOil
UNREFINED Oil DESTINED FOR HUMAN CONSUMPTION (79.5% of Production)
/ p Stwqie Tanks
Unrefined Oil
REFINED [ 20~of OIL ~, Production ]
t t
Deodourization
IReachlng
t Wuhln s
lieflnlnll
t t
7- (311Dqqpimml~
Fig. 2. Rapeseed oil production, company B, 1987.
consumption, inspected trucks were not used. According to the buyers' requirements crude rapeseed oil is denatured with 2% aniline or with ceres blue or ricin oil at a concentration of 5%. At the request of the Spanish buyer, the oil was denatured with 2% aniline before shipment. The aniline was standard technical-grade material which assayed >99.8% (by gas chromatography), according to the specification sheet retained in company A's records. Aniline requirements were calculated on a 2% weight/weight basis. Aniline was placed in the tanks before the oil, and the two were assumed to have mixed during loading and transportation. Since the capacity of the tanks was about 20 tonnes each, about 20 to 25 truckloads were individually mixed in this manner in 1981. In 1981, company A sold about 500 metric tonnes (about 3 days of production) of rapeseed oil to Spain for industrial use. No criteria were used to select oil for industrial use from all the crude oil produced. Some oil sold to Spain was partially refined by company A; however, after oil enters the industrialoil refining section of the factory it does not return to the food-oil section. Although the partially refined denatured rapeseed oil was sold to Spanish companies on four occasions, on only one occasion was aniline used as denaturant. Only low erucic acid rapeseed varieties were available to this factory in 1981, and the factory administrators considered that the diversion of oil low in erucic acid to industrial use was unremarkable.
Company B--general description On 12 November 1987, we visited company B, which is smaller than company A. It is a family-run
company, dedicated primarily to the production of rapeseed oil, almost all of which is used for human consumption; however, walnut, hazelnut and almond oils are also produced in a building separate from those where rapeseed oil is processed. Company B operates 6 days a week, stopping all activity every Sunday. The facility was clean and orderly. Factory capacity is about 50 metric tonnes of oil per day, of which about 40 tonnes is sold directly as crude oil, without degumming, for further refining by the company's customers. About 10 tonnes of rapeseed oil per day is refined and sold directly for human consumption. Almost all rapeseed oil produced at the plant is eventually used for human consumption, and only about 0.5% for industrial use. Some oil for human consumption is bottled at the factory; the remainder is pumped into tank trucks for shipment to customers. The overall operation is basically the same as it was in 1981.
Company B--seed Seeds arrive at this factory by truck, generally standard 25-tonne-capacity trucks, although occasionally special tankers designed for the transport of food grains are used. All of the seeds processed at this location are grown in France within about a 100-mile radius of the factory; they may be obtained directly from farms, co-operatives, or seed storage companies. Before reaching the plant, seeds are stored for, at most, 11 months. The amount of time seed is held before being sent to the plant depends on the time of year in which it is processed, since rapeseed is generally harvested in June to August of
Rapeseed oil manufacture and TOS each year. The proprietors believe that the rapeseed processed in 1981 was of the Jet Neuf variety. On delivery the seed is inspected directly for appearance and odour and analysed in the laboratory for oil and moisture content. The seed is then stored, as necessary, in vertical metal silos of corrugated iron, enclosed within a concrete warehouse. The storage area is clean, and stainless-steel is used extensively. Seed is dried and aerated during storage but is not treated with any chemical preservative. The storage capacity at this facility is about 900-1000 tonnes of seed. Since about 500-600 tonnes of seed are processed each week, the maximum storage time is about 15 days. Normally, however, seed is stored for no more than 8 days.
Company B--crude oil production The manufacturing process is continuous. The seed is milled (flaked) and heated to 100°C before it is pressed in four screw presses. The press cake is sent continuously to be extracted by a vertical extractor. About 0.5 litres of hexane per tonne of cake are lost in the solvent extraction process. Solvent is then removed from the degreased cake, after which the cake is dried and stored. The oil is stored in vertical tanks (Fig. 2). Company B--oil refining Oil is refined by the following discontinuous process: (1) degumming with water, (2) neutralizing, (3) washing, (4) bleaching, and (5) deodorizing. The oil is neutralized in a vertical tank with caustic soda at 20 ° Baum~; it is bleached with bentonite, filtered with two press filters, and deodorized at 190°C and 3 mmHg (pressure). The factory's refining capacity is 10 tonnes per day (Fig. 2). Company B never sold refined rapeseed oil to Spanish companies. Company B--denaturation In 1981 oil was denatured at company B by mixing aniline and unrefined rapeseed oil in a 10-tonne capacity tank. This tank had been used for ricin oil but had been out of service for several years before 1981. The pump used for aniline had previously been used to pump fuel oil; however, it was cleaned with steam and caustic soda before being used for aniline. Another pump was used to move the denatured oil from this tank to the tank trucks. When the company stopped selling denatured rapeseed oil, both pumps were destroyed as well as the pipes that had been used. The tank trucks are inspected before the products are loaded. No specific certification requires that the trucks be cleaned; however, company B insists that the tank trucks be steam cleaned before its oil is loaded. DISCUSSION
Despite extensive investigation of all the steps involved in transporting, handling and refining of the aniline denatured rapeseed oil after its entry into Spain, the point at which the causative agent of TOS was introduced has not been determined. Several hypotheses have been offered regarding the possible identity of the aetiological agent (Bernert et al., 1989;
801
Kammiiller et al., 1988; V~quez Roncero et al., 1984) but none can be considered proven. Since aniline was the denaturant used in the suspect oil (and therefore the primary contaminant), aniline and its reaction products with normal components of edible oils have received particular attention as possible aetiological agents (Bernert et al., 1987 and 1989; Grandjean and Tarkowski, 1984; Kilbourne et al., 1988). However, the clinical manifestations of TOS are not those of aniline toxicity (Grandjean and Tarkowski, 1984; Tabuenca, 1981). Moreover, despite the strong correlation between the concentration of certain aniline-containing compounds (fatty acid anilides) and the risk of illness associated with consumption of a particular oil, these compounds may serve only as markers for the true aetiological agent. The lack of certainty about the nature of the causative agent of TOS makes it imperative to explore all possible contaminants and modes of contamination including those possibly related to the production and handling of the oil before its importation into Spain. In this paper, we have reviewed the process for producing denatured rapeseed oil at the two French companies involved. Both companies bought French seed of quality sufficient to qualify the company for European Common Market subsidies. At company B, the owners believed the genetic variety of seed used was Jet Neuf; but at company A the exact type used was unknown. Nevertheless, personnel from both companies stated that high erucic acid seeds were not available in France in 1981. Seeds were transported to the factories by the usual methods, which did not appear to present any risk of contamination. In addition, no record was kept of any special lot of seeds from which oil was designated for Spain. Seed storage at both factories was generally relatively short (one month or less), although seeds may have been stored before shipment to the factories. We found no major differences in crude oil production at the two companies. Neither factory degummed the crude oil before sale. However, the finding that French companies do not degum crude oil may be an important difference from the way Spanish companies produce oil. Possibly, the Spanish companies were expecting to receive a product that had been degummed, although not further refined. Most oil from both companies was sold to French refining companies for further processing before its sale for human consumption. The oil refined by company A was for industrial use and therefore not deodorized. Although company A sold this partially refined denatured rapeseed oil to Spanish companies on four occasions, on only one occasion was the oil denatured with aniline. Company B never sold refined rapeseed oil to Spanish companies. The TOS epidemic was probably not a result of a defective or unhealthy rapeseed oil product produced at one of the two oil extraction plants visited. Most rapeseed oil produced in both plants in 1981 was for human consumption (99.5% in company B and 98% in company A). According to the information provided by the companies, all rapeseed oil sent to Spain in 1981 would have been fit for human consumption had it undergone the usual refining procedures and not been denatured. To support this
802
M.
POSADA
DE LA PAZ
argument, both companies mentioned that product taken from the same stocks as the product sold to the Spanish was sold on the French market for human consumption without any apparent adverse effect on health. Company A personnel made the point that as soon as their oil was denatured with aniline, they no longer considered the product fit for human consumption, and the customary standards of cleanliness and protection from chemical contamination in storage and transportation were not necessarily applied. Company A officials believed that the trucking companies that took the oil to Spain usually transported chemical products and that the tank trucks were not necessarily washed well enough to rid them of all traces of their previous contents before rapeseed oil was loaded. Company A officials did note, however, that their usual practice would have been to visually inspect the tank trucks to make sure they were clean and dry before their product was loaded. In 1981, virtually all rapeseed oil entering Spain from these two factories was undegummed, unrefined oil with 2% aniline added as a denaturant. A few batches of oil from company A were partially refined before they were denatured and sent to Spain. Only one of these batches was denatured with aniline and sent to a Spanish company involved in the epidemic. Aniline treatment may have caused different compounds to form in the oil in this batch. Therefore, this particular batch of oil is of special interest. Oil was never denatured for sale to non-Spanish companies. Both companies A and B denatured the rapeseed oil sold to Spanish companies because they understood that this process was required by Spanish law. The denaturant was changed to aniline from ricin or ceres blue at the request of a Spanish oil importer and this was interpreted by the French companies as being due to a change in Spanish law. In addition, at company B, aniline and oil were mixed in containers not used for other purposes; at company A, they were mixed directly in the tank trucks, leaving little chance for aniline contamination of oil that was not sold to Spain. According to information from the Spanish customs authorities, company A had been selling rapeseed oil denatured with 2% aniline to Spanish companies since 1979 or possibly earlier. Whether that oil was eventually used for human consumption or industrial use is not known. Two clear differences existed between oils sold by the same companies to Spanish importers and oils sold in France for domestic consumption: (1) the 2% aniline denaturant in the oils sent to Spain and (2) the possibility that the Spanish oils were further contaminated by unknown agents in the added aniline or in the tank trucks. Although an aniline derivative or a contaminant of the technical-grade aniline used may be the causative agent of TOS, the possibility that TOS was caused by unidentified contaminants in tank trucks that were not appropriately prepared to carry food deserves serious consideration. In addition, .our data indicate that we should consider the possibility that TOS was caused by a reaction product of normal oil components or of the materials used in oil refining with a contaminant in technical-grade aniline or a transport-related contaminant.
et
aI.
The Spanish government has attempted to obtain records from the trucking companies that transported the denatured rapeseed oil to Spain. Specifically, it has requested, through diplomatic and administrative channels, data about the types of products carried in the tank trucks immediately before they carried the denatured rapeseed oil and data about the materials usually transported by these trucks. This information could help determine how any transportation-related chemical contamination might have caused TOS.
REFERENCES
Bemert J. T., Jr, Kilboume E. M., Akins J. R., Posada de la Paz M., Meredith N. K., Abaitua Borda I. and Wages S. (1987) analvsis of oil samules . _Compositional _ . imolicated . in the Spanish toxic oil syndrome. Journal of Food Science 52, 1562-1569. Bernet? J. T., Jr, Pendergrast A. H., Ashley D. L., Patterson D. G., Jr, Kilbourne E. M., Alexander L. R., Posada de la Paz M. and Abaitua Borda I. (1989) Synthesis of N-(5-vinvl- 1.3-thiazolidin-2-vlidenemhenvlamine and -. analysis of bils.implicated in the Spanish toxic oil syndrome for its presence. Food and Chemical Toxicology 27, 159164.
Cafias R. and Kilbourne E. M. (1987) Oil ingestion and the toxic-oil syndrome: results of a survey of residents of the Orcasur neighbourhood in Madrid, Spain. International Journal of Epidemiology 16, 36.
Casado Flores J., Casquero J., Colomar P., Garcia Ferrer0 M., Morales M., Sanchez Palacin A. and Tabuenca J. M. (1982) Sindrome toxic0 por consume de aceite adulterado. Ciclo Ensayos Medicos 31, 9-12. Diaz de Rojas F., Castro Garcia M., Abaitua Borda I., Alonso Gordo J. M., Posada de la Paz M., Kilbourne E. M. and Tabuenca Oliver J. M. (1987) The association of oil ingestion with toxic-oil syndrome in two convents. American Journal of Epidemiology 125, 907-911.
Equip0 Especial de Investigation, Comisi6n Interministerial para el Asunto de1 Problema de1 Aceite de Colza Desnaturalizado. (1981) Infome de1 Equip0 Especial: Investigacian de1 aceite a’e colza desnaturalizado. Volumen I.
Ministerio de Trabajo, Sanidad y Seguridad Social; Ministerio de Agricultura y Pesca; Ministerio de Economia Y Comercio: Ministerio de1 Interior: and Ministerio de Industria y Energia, Madrid. Gardner A. M., Yurawecz M. P., Cunningham W. C., Diachenko G. W., Mazzola E. P. and Brumley W. C. (1983) Isolation and identification of Cl6 and Cl8 fatty acid esters of chloropropanediol in adulterated Spanish cooking oils. Bulletin of Environmental Contamination and Toxicology 31, 625430.
P. and Tarkowski S. (Editors) (1984) Toxic Oil Syndrome. Mass Food Poisoning in Spain. Report of a WHO Meeting: Madrid 21-25 March 1983. World
Grandjean
Health Organization Regional Office for Europe, Copenhagen. Hill R. H., Jr, Todd G. D., Kilboume E. M., Cline R. E., McGraw J., Orti D. L., Bailey S. L. and Needham L. L. (1987) Gas chromatographic/mass spectrometric determination of aniline in food oils associated with the Soanish toxic-oil syndrome. Bulletin of Environmental Contamination and Toxicology 39, 51 l-515.
Kammiiller M. E., Verhaar H. J. M., Versluis C., Terlouw J. K., Brandsma L., Penninks A. H. and Seinen W. (1988) I-Phenyl-5-vinyl-2-imidazolidinethione, a proposed causative agent of Spanish toxic oil syndrome: synthesis, and identification in one of a group of case-associated samples. Food and Chemical Toxicology 26, 119127.
Rapeseed oil manufacture and TOS Kilbourne E. M., Bernert J. T., Jr, Posada de la Paz M., Hill R. H., Jr, Abaitua Borda I., Kilbourne B. W., Zack M. M. and The Toxico-Epidemiologic Study Group. (1988) Chemical correlates of pathogenicity of oils related to the toxic-oil syndrome in Spain. American Journal of Epidemiology 127, 1210-1227. Posada M., Castro M., Kilbourne E. M., Diaz de Rojas F., Abaitua I., Tabuenca J. M. and Vioque A. (1987) Toxicoil syndrome case reports associated with the ITH oil refinery in Sevilla. Food and Chemical Toxicology 25, 87-90. Rigau-P6rez J. G., P6rez-Alvarez L., Duefias-Castro S., Choi K., Thacker S. B., Germain J. L., Gonz/dez de Andr6s G., Cafiada Royo L. and P6rez Gallardo F. (1984)
803
Epidemiologic investigation of an oil-associated pneumonic paralytic eosinophilic syndrome in Spain. American Journal of Epidemiology 119, 250-260. Tabuenca J. M. (1981) Toxic allergic syndrome caused by ingestion of rapeseed oil denatured with aniline. Lancet ii, 567-568. V~zquez Roncero A., Maestro Duran R. and Ruiz Guti6rrez V. (1984) New aniline derivatives in "toxic oil syndrome": toxicity in mice of 3-phenylamino-l,2propanediol and its fatty acid mono- and diesters. Grasas y Aceites 35, 330-331. Vioque E. and Vioque A. (1983) Presence of peroxidized fatty anilides in oils purportedly provoking toxic syndrome. Grasas y Aceites 34, 159-161.
0278-6915/91 $3.00 + 0.00 Pergamon Press plc
Printed in Great Britain
Research Section MANUFACTURING PROCESSES AT TWO FRENCH RAPESEED OIL COMPANIES: POSSIBLE RELATIONSHIPS TO TOXIC OIL SYNDROME IN SPAIN M. POSADADE L A PAZ*, R. M. PHILENt:~, I. ABAITUABORDA*,J. T. BERNERT,JR:~, J. C. BADAGANCEDO~,P. J. DUCLOS~,[[ and E. M. KILBOURNE:~ *Unidad de Programas de Investigaci6n, Fondo de Investigaci6n Sanitaria, Ministerio de Sanidad y Consumo, Antonio Grilo 10, 28015-Madrid, Spain, ~/Centersfor Disease Control, Public Health Service, US Department of Health and Human Services, Atlanta, GA 30333, USA, §Director del Instituto de Productos, Lficteos de Asturias, Apartado de Correos 85, 33300 Villaviciosa, Asturias, Spain and l[Laboratory Centre for Disease Control, Health and Welfare Canada, Tunney'sPasture, Ottawa, Ontario K1A 0L2, Canada (Accepted 29 July 1991) Abstract--The toxic oil syndrome (TOS) epidemicthat occurred in Spain in spring 1981 has been associated with the consumption of rapeseed oil that was denatured with aniline for industrial use but diverted for human consumption. The precise aetiologic agent in the oil responsible for the outbreak has not been identified. To learn more about possible contaminants and how the contamination might have occurred, we visitedtwo French companies that process rapeseed oil and that were identifiedin Spanish administrative and judicial records as the ones exporting aniline-denatured rapeseed oil to Spain in 1981. With the apparently full and voluntary co-operation of personnel at both companies, we reviewed the processes involvedin manufacturing, treating and transporting rapeseed oil, and we have summarizedthe information provided to us. Of particular importance is the finding that oil exported to Spain was taken from stock, the rest of which was sold for human consumption in the French domestic market, apparently without any adverse health effects. The differences between the oil exported to Spain and the oil sold as food in France were that aniline equivalent to 2% of the weight of the oil was added to most of the Spanish oil but not to that sold in France, and that contamination of the Spanish oil may have occurred in the tank trucks used for transportation to Spain, which had previously carried industrial chemicals. There is no assurance that the trucks were cleaned appropriately for transporting a food product before the oil was loaded for the journey to Spain. Since the clinical manifestations of TOS are not those of aniline toxicity, we conclude that the aetiological agent of TOS is likely to be one of the following: (1) a contaminant in the aniline, (2) a contaminant introduced during transportation, (3) a reaction product of normal oil components or materials used in refiningwith either aniline or the potential contaminants mentioned under (1) or (2) above.
INTRODUCTION
Toxic oil syndrome (TOS) was a previously unrecognized illness that occurred in Spain in epidemic form in spring 1981; it affected more than 20,000 people and caused several hundred deaths (Grandjean and Tarkowski, 1984). Epidemiological investigations have shown that this outbreak was caused by the ingestion of aniline-denatured rapeseed oil initially intended for industrial use, but diverted to human consumption (Cafias and Kilbourne, 1987; Casado Flores et al., 1982; Diaz de Rojas et al., 1987; Kilbourne et al., 1988; Posada et al., 1987; RigauPtrez et al., 1984; Tabuenca, 1981). Although several contaminants have been identified in implicated oils, the precise aetiological agent responsible for TOS
tTo whom all correspondence should be addressed, except reprint requests, which should be addressed to Dr Posada de la Paz. Abbreviations: TPS -- toxic oil syndrome. FCT 29/12--A
has yet to be identified (Bernert et al., 1987; Gardner et al., 1983; Grandjean and Tarkowski, 1984; Hill et al., 1987; Vioque and Vioque, 1983). The transportation and handling of the suspect oil after its importation into Spain has been investigated in detail (Equipo Especial de Investigaci6n, 1981); however, the manufacture and handling of the oil in France has not been described previously. Our review of Spanish administrative and judicial records identified two French companies as the only sources of aniline-denatured rapeseed oil exported to Spain in 1981. We therefore visited these companies to review all steps in the manufacture, treatment and transportation of the implicated oils and to identify any aspects of these procedures that might have led to contamination of oil with the substance or substances that cause TOS. MATERIAI~ AND METHODS
Two of us (M.P. and I.A.) reviewed the six volumes of information that have been collected by Spanish
797
798
M. POSADADE LA PAZ et al.
administrative and judicial authorities regarding the importation of rapeseed oils into Spain in 1981 and previous years. We abstracted data on (1) dates when specific oil shipments left France and entered Spain between January and June 1981; (2) type of oil (whether denatured, the type of denaturant used, and whether refined); (3) quantity of oil; (4) licence numbers of tank trucks used; and (5) other information. We collected data in the same way at the two French companies (companies A and B). At both factories we met with the administrative directors and their associates, as well as with technical personnel directly involved in oil production. We thoroughly inspected the oil production facility and techniques. In particular, we investigated the techniques for adding aniline. We also investigated other possible routes for oil contamination, and typical production runs. The type of information we attempted to elicit is listed below: General activities of the companytypes of activities, volume of production, administrative or technical changes since 1981, products used; seedgenetic variety, geographic origin, content of erucic acid; handling of the seedtransport, quality inspection at arrival, unloading, cleaning, drying, storage; oil processingpressing, solvent extraction, degumming, refining, storage of oil; oil transportcleaning of tankers, transportation companies, final destination(s) of the oils; denaturantscompounds used and chemical characteristics, reasons for changing from one compound to another, governmental regulations governing use of denaturants, mixture of the denaturants with oil, companies involved in the manufacture of denaturants; movement of specific lots of oilnumbers of lots shipped, weights of lots shipped, specific trucks used, shipment dates; Factory operationstypes of machinery present, processing details, temperature, pressure, etc., maintenance of the installations/machinery.
Both companies gave us product samples, including rapeseed oil (crude and refined, if both were available). All samples were of recent origin, since samples from 1981 were not available. RESULTS
In general, the invoices and other records of the two companies agreed well with the data obtained from the Spanish customs authorities. These data related to shipping dates, quantities of denatured oil shipped and the destinations to which the oil was sent. Extracting crude oil from rapeseed is a continuous process with the following steps: (1) cleaning and storing, (2) milling, (3) heating, (4) pressing, and (5) extracting with solvent. The details of these processes are discussed separately for each company. Company A-general
description
On 10 November 1987 we visited company A, which is a subsidiary of a larger holding company that has changed ownership since 198 1. The previous holding company went into bankruptcy for reasons unrelated to the sale of rapeseed oil to Spain. Despite the change, the staff and activities of the company remain the same as in 1981. Company A has three departments that each perform different activities. One department produces and sells crude (unrefined) vegetable oil that is transported in tank trucks; this oil is refined by the buyers and then sold principally for human consumption. The second department produces refined oil for industrial use, and the third department produces resin. Normally, the activities of the three departments are strictly segregated. Only the activities of the first two departments are described in detail here. The factory processes both rapeseed and sunflower seed according to the dictates of the parent (holding) company. It has the capacity to process daily 450 metric tonnes of seed, which produces 180 to 200 metric tonnes of oil a day. According to the technical managers, 98% of the oil is for human consumption, and 2% is for industrial use. The operating facilities and procedures are basically the same as in 1981, when they sold about 500 metric tonnes of rapeseed oil to Spain for industrial use. Company A-seed
The rapeseed is of genetic varieties low in erucic acid and comes to the factory from all over France by truck or rail. Upon arrival, the seed is inspected for quality: the colour and odour is examined, the oil and moisture content is determined, and the oil acidity and erucic acid content is measured. At the factory, rapeseed from different regions and of different varieties are mixed; therefore oil cannot be traced through production records to seed from a particular location or of a specific variety. The plant operators stated that they have no interest in such factors; their concerns are mainly with the oil content and the level of erucic acid. In 1981, the erucic acid content was less than 5%. Although no tests are performed for specific biological or chemical contaminants, no known problems with pesticide contamination have occurred.
799
Rapeseed oil manufacture and TOS Salefur IndustrialUae
Seed f ~ r ~
t
1Ildl~
(~ines)
t
Conk3 1oo*c 15-30 rain
~
t
Sale for Human Consumption
Water Wash and Vacuum Dr)'
t t ~d.
L
Bleached
t
Neutralizationwith Caustic Soda
t
Oepnlmed w~h
t
Ph&pho~c~d
Shipment
Animal Feed Stool Oil (Mixed laddsH)
l
Refined at > CompanyA
Fig. 1. Rapeseed oil production, company A, 1987. After inspection, the seed is unloaded and filtered to separate out stones, pieces of metal, and other foreign material, and is then conveyed to concrete storage silos where it remains for at most 15 days before further processing. There are no facilities for long-term seed storage at the location we visited.
Company A---crude oil production Crude oil is prepared by a standard process, as shown in Fig. 1. Seed is transported from the storage silos to crushing rollers (flaking machines) and then to cookers, which were in operation at about 100°C on the day of our visit. Although sunflower seed was being processed when we were at the factory, we were told that the same conditions (100°C for 15-30 min) are normally used for rapeseed. Cooked seeds are pressed in screw-press expellers, and the expressed oil is then centrifuged, filtered and stored. The press cake from the expellers is conveyed to a separate building where additional oil is extracted with hexane (the only solvent Company A has ever used) by a standard continuous counter-current D'smet solvent-extraction process that has not been altered since 1981. After the hexane is recovered, the extracted oil is stored, often in a mixture with an earlier batch of pressed oil. Only crude oil is prepared by this process; the oil is not degummed or treated in any other way before storage and eventual shipment. The overall crude oil preparation operates continuously, 24 hr a day, 7 days a week, except for *The term used by oil producers to define the concentration of sodium hydroxide used. For liquids heavier than water (such as sodium hydroxide solutions) the Baum~ scale is based on the density of a 10% sodium ehioride solution, which is given a value of 10° Baum~ and the density of pure water which is given a value of 0° Baum6.
a yearly shutdown for general maintenance, usually in July. The crude oil is sold and delivered in tank trucks and trains to refineries that process the oil further before it is sold for human consumption. Before shipment, oil is stored for periods ranging from a few days to 1 month in 1000-tonne capacity tanks. After the solvent extraction process, the remaining press cake is sold as animal feed.
Company A--oil refining At company A, refining oil is a much smaller operation than extracting oil. The oils refined include rapeseed, sunflower, linseed and soybean. Whereas almost all the crude oil produced is intended for human consumption after further refining elsewhere, the refined oil is strictly for industrial use. The refining process used is a common discontinuous (batch) process, as shown in Fig. 1. The steps in refining are as follows: (1) degumming with phosphoric acid, (2) neutralizing with caustic soda (18 ° Baum~*) in four vertical tanks of 15 tonnes, each with gravimetric separation, (3) washing with water and drying under vacuum, and (4) decolourizing (bleaching) with bleaching earth normally obtained from Germany, although an English clay is used occasionally. Spent bleaching earth is removed from the oil by a standard filter press. All bleaching earths used in this process are identical to those used for processing food oils. The industrial oils are not deodorized, hydrogenated, or otherwise modified.
Company A-~lenaturation Tank trucks are inspected and then loaded for shipment to buyers. Trucks that carry consumable oil have inspection cards valid for all of Europe. F o r denatured rapeseed oil, not intended for human
800
M. POSADADE LA PAZ et al. rm,~ x~.,eed ~m, Flum
SeedStorap Co-operatives Companies Steam.Clewed Tanker Trucks Transportby Truck or r~od TmKen
( o.sof
Production]
DENATURED OIL Aniline Added
Storage (Dried and Aerated)
Process 10-TonTankfor Oeanturization
MIIIInl (Flaking) Hexlne
\ Dried Press 4 Cake Stered
HexaneExtraction
t
4 Screw Presses
1
Ip PreooCake
1
UnrefinedOil
UNREFINED Oil DESTINED FOR HUMAN CONSUMPTION (79.5% of Production)
/ p Stwqie Tanks
Unrefined Oil
REFINED [ 20~of OIL ~, Production ]
t t
Deodourization
IReachlng
t Wuhln s
lieflnlnll
t t
7- (311Dqqpimml~
Fig. 2. Rapeseed oil production, company B, 1987.
consumption, inspected trucks were not used. According to the buyers' requirements crude rapeseed oil is denatured with 2% aniline or with ceres blue or ricin oil at a concentration of 5%. At the request of the Spanish buyer, the oil was denatured with 2% aniline before shipment. The aniline was standard technical-grade material which assayed >99.8% (by gas chromatography), according to the specification sheet retained in company A's records. Aniline requirements were calculated on a 2% weight/weight basis. Aniline was placed in the tanks before the oil, and the two were assumed to have mixed during loading and transportation. Since the capacity of the tanks was about 20 tonnes each, about 20 to 25 truckloads were individually mixed in this manner in 1981. In 1981, company A sold about 500 metric tonnes (about 3 days of production) of rapeseed oil to Spain for industrial use. No criteria were used to select oil for industrial use from all the crude oil produced. Some oil sold to Spain was partially refined by company A; however, after oil enters the industrialoil refining section of the factory it does not return to the food-oil section. Although the partially refined denatured rapeseed oil was sold to Spanish companies on four occasions, on only one occasion was aniline used as denaturant. Only low erucic acid rapeseed varieties were available to this factory in 1981, and the factory administrators considered that the diversion of oil low in erucic acid to industrial use was unremarkable.
Company B--general description On 12 November 1987, we visited company B, which is smaller than company A. It is a family-run
company, dedicated primarily to the production of rapeseed oil, almost all of which is used for human consumption; however, walnut, hazelnut and almond oils are also produced in a building separate from those where rapeseed oil is processed. Company B operates 6 days a week, stopping all activity every Sunday. The facility was clean and orderly. Factory capacity is about 50 metric tonnes of oil per day, of which about 40 tonnes is sold directly as crude oil, without degumming, for further refining by the company's customers. About 10 tonnes of rapeseed oil per day is refined and sold directly for human consumption. Almost all rapeseed oil produced at the plant is eventually used for human consumption, and only about 0.5% for industrial use. Some oil for human consumption is bottled at the factory; the remainder is pumped into tank trucks for shipment to customers. The overall operation is basically the same as it was in 1981.
Company B--seed Seeds arrive at this factory by truck, generally standard 25-tonne-capacity trucks, although occasionally special tankers designed for the transport of food grains are used. All of the seeds processed at this location are grown in France within about a 100-mile radius of the factory; they may be obtained directly from farms, co-operatives, or seed storage companies. Before reaching the plant, seeds are stored for, at most, 11 months. The amount of time seed is held before being sent to the plant depends on the time of year in which it is processed, since rapeseed is generally harvested in June to August of
Rapeseed oil manufacture and TOS each year. The proprietors believe that the rapeseed processed in 1981 was of the Jet Neuf variety. On delivery the seed is inspected directly for appearance and odour and analysed in the laboratory for oil and moisture content. The seed is then stored, as necessary, in vertical metal silos of corrugated iron, enclosed within a concrete warehouse. The storage area is clean, and stainless-steel is used extensively. Seed is dried and aerated during storage but is not treated with any chemical preservative. The storage capacity at this facility is about 900-1000 tonnes of seed. Since about 500-600 tonnes of seed are processed each week, the maximum storage time is about 15 days. Normally, however, seed is stored for no more than 8 days.
Company B--crude oil production The manufacturing process is continuous. The seed is milled (flaked) and heated to 100°C before it is pressed in four screw presses. The press cake is sent continuously to be extracted by a vertical extractor. About 0.5 litres of hexane per tonne of cake are lost in the solvent extraction process. Solvent is then removed from the degreased cake, after which the cake is dried and stored. The oil is stored in vertical tanks (Fig. 2). Company B--oil refining Oil is refined by the following discontinuous process: (1) degumming with water, (2) neutralizing, (3) washing, (4) bleaching, and (5) deodorizing. The oil is neutralized in a vertical tank with caustic soda at 20 ° Baum~; it is bleached with bentonite, filtered with two press filters, and deodorized at 190°C and 3 mmHg (pressure). The factory's refining capacity is 10 tonnes per day (Fig. 2). Company B never sold refined rapeseed oil to Spanish companies. Company B--denaturation In 1981 oil was denatured at company B by mixing aniline and unrefined rapeseed oil in a 10-tonne capacity tank. This tank had been used for ricin oil but had been out of service for several years before 1981. The pump used for aniline had previously been used to pump fuel oil; however, it was cleaned with steam and caustic soda before being used for aniline. Another pump was used to move the denatured oil from this tank to the tank trucks. When the company stopped selling denatured rapeseed oil, both pumps were destroyed as well as the pipes that had been used. The tank trucks are inspected before the products are loaded. No specific certification requires that the trucks be cleaned; however, company B insists that the tank trucks be steam cleaned before its oil is loaded. DISCUSSION
Despite extensive investigation of all the steps involved in transporting, handling and refining of the aniline denatured rapeseed oil after its entry into Spain, the point at which the causative agent of TOS was introduced has not been determined. Several hypotheses have been offered regarding the possible identity of the aetiological agent (Bernert et al., 1989;
801
Kammiiller et al., 1988; V~quez Roncero et al., 1984) but none can be considered proven. Since aniline was the denaturant used in the suspect oil (and therefore the primary contaminant), aniline and its reaction products with normal components of edible oils have received particular attention as possible aetiological agents (Bernert et al., 1987 and 1989; Grandjean and Tarkowski, 1984; Kilbourne et al., 1988). However, the clinical manifestations of TOS are not those of aniline toxicity (Grandjean and Tarkowski, 1984; Tabuenca, 1981). Moreover, despite the strong correlation between the concentration of certain aniline-containing compounds (fatty acid anilides) and the risk of illness associated with consumption of a particular oil, these compounds may serve only as markers for the true aetiological agent. The lack of certainty about the nature of the causative agent of TOS makes it imperative to explore all possible contaminants and modes of contamination including those possibly related to the production and handling of the oil before its importation into Spain. In this paper, we have reviewed the process for producing denatured rapeseed oil at the two French companies involved. Both companies bought French seed of quality sufficient to qualify the company for European Common Market subsidies. At company B, the owners believed the genetic variety of seed used was Jet Neuf; but at company A the exact type used was unknown. Nevertheless, personnel from both companies stated that high erucic acid seeds were not available in France in 1981. Seeds were transported to the factories by the usual methods, which did not appear to present any risk of contamination. In addition, no record was kept of any special lot of seeds from which oil was designated for Spain. Seed storage at both factories was generally relatively short (one month or less), although seeds may have been stored before shipment to the factories. We found no major differences in crude oil production at the two companies. Neither factory degummed the crude oil before sale. However, the finding that French companies do not degum crude oil may be an important difference from the way Spanish companies produce oil. Possibly, the Spanish companies were expecting to receive a product that had been degummed, although not further refined. Most oil from both companies was sold to French refining companies for further processing before its sale for human consumption. The oil refined by company A was for industrial use and therefore not deodorized. Although company A sold this partially refined denatured rapeseed oil to Spanish companies on four occasions, on only one occasion was the oil denatured with aniline. Company B never sold refined rapeseed oil to Spanish companies. The TOS epidemic was probably not a result of a defective or unhealthy rapeseed oil product produced at one of the two oil extraction plants visited. Most rapeseed oil produced in both plants in 1981 was for human consumption (99.5% in company B and 98% in company A). According to the information provided by the companies, all rapeseed oil sent to Spain in 1981 would have been fit for human consumption had it undergone the usual refining procedures and not been denatured. To support this
802
M.
POSADA
DE LA PAZ
argument, both companies mentioned that product taken from the same stocks as the product sold to the Spanish was sold on the French market for human consumption without any apparent adverse effect on health. Company A personnel made the point that as soon as their oil was denatured with aniline, they no longer considered the product fit for human consumption, and the customary standards of cleanliness and protection from chemical contamination in storage and transportation were not necessarily applied. Company A officials believed that the trucking companies that took the oil to Spain usually transported chemical products and that the tank trucks were not necessarily washed well enough to rid them of all traces of their previous contents before rapeseed oil was loaded. Company A officials did note, however, that their usual practice would have been to visually inspect the tank trucks to make sure they were clean and dry before their product was loaded. In 1981, virtually all rapeseed oil entering Spain from these two factories was undegummed, unrefined oil with 2% aniline added as a denaturant. A few batches of oil from company A were partially refined before they were denatured and sent to Spain. Only one of these batches was denatured with aniline and sent to a Spanish company involved in the epidemic. Aniline treatment may have caused different compounds to form in the oil in this batch. Therefore, this particular batch of oil is of special interest. Oil was never denatured for sale to non-Spanish companies. Both companies A and B denatured the rapeseed oil sold to Spanish companies because they understood that this process was required by Spanish law. The denaturant was changed to aniline from ricin or ceres blue at the request of a Spanish oil importer and this was interpreted by the French companies as being due to a change in Spanish law. In addition, at company B, aniline and oil were mixed in containers not used for other purposes; at company A, they were mixed directly in the tank trucks, leaving little chance for aniline contamination of oil that was not sold to Spain. According to information from the Spanish customs authorities, company A had been selling rapeseed oil denatured with 2% aniline to Spanish companies since 1979 or possibly earlier. Whether that oil was eventually used for human consumption or industrial use is not known. Two clear differences existed between oils sold by the same companies to Spanish importers and oils sold in France for domestic consumption: (1) the 2% aniline denaturant in the oils sent to Spain and (2) the possibility that the Spanish oils were further contaminated by unknown agents in the added aniline or in the tank trucks. Although an aniline derivative or a contaminant of the technical-grade aniline used may be the causative agent of TOS, the possibility that TOS was caused by unidentified contaminants in tank trucks that were not appropriately prepared to carry food deserves serious consideration. In addition, .our data indicate that we should consider the possibility that TOS was caused by a reaction product of normal oil components or of the materials used in oil refining with a contaminant in technical-grade aniline or a transport-related contaminant.
et
aI.
The Spanish government has attempted to obtain records from the trucking companies that transported the denatured rapeseed oil to Spain. Specifically, it has requested, through diplomatic and administrative channels, data about the types of products carried in the tank trucks immediately before they carried the denatured rapeseed oil and data about the materials usually transported by these trucks. This information could help determine how any transportation-related chemical contamination might have caused TOS.
REFERENCES
Bemert J. T., Jr, Kilboume E. M., Akins J. R., Posada de la Paz M., Meredith N. K., Abaitua Borda I. and Wages S. (1987) analvsis of oil samules . _Compositional _ . imolicated . in the Spanish toxic oil syndrome. Journal of Food Science 52, 1562-1569. Bernet? J. T., Jr, Pendergrast A. H., Ashley D. L., Patterson D. G., Jr, Kilbourne E. M., Alexander L. R., Posada de la Paz M. and Abaitua Borda I. (1989) Synthesis of N-(5-vinvl- 1.3-thiazolidin-2-vlidenemhenvlamine and -. analysis of bils.implicated in the Spanish toxic oil syndrome for its presence. Food and Chemical Toxicology 27, 159164.
Cafias R. and Kilbourne E. M. (1987) Oil ingestion and the toxic-oil syndrome: results of a survey of residents of the Orcasur neighbourhood in Madrid, Spain. International Journal of Epidemiology 16, 36.
Casado Flores J., Casquero J., Colomar P., Garcia Ferrer0 M., Morales M., Sanchez Palacin A. and Tabuenca J. M. (1982) Sindrome toxic0 por consume de aceite adulterado. Ciclo Ensayos Medicos 31, 9-12. Diaz de Rojas F., Castro Garcia M., Abaitua Borda I., Alonso Gordo J. M., Posada de la Paz M., Kilbourne E. M. and Tabuenca Oliver J. M. (1987) The association of oil ingestion with toxic-oil syndrome in two convents. American Journal of Epidemiology 125, 907-911.
Equip0 Especial de Investigation, Comisi6n Interministerial para el Asunto de1 Problema de1 Aceite de Colza Desnaturalizado. (1981) Infome de1 Equip0 Especial: Investigacian de1 aceite a’e colza desnaturalizado. Volumen I.
Ministerio de Trabajo, Sanidad y Seguridad Social; Ministerio de Agricultura y Pesca; Ministerio de Economia Y Comercio: Ministerio de1 Interior: and Ministerio de Industria y Energia, Madrid. Gardner A. M., Yurawecz M. P., Cunningham W. C., Diachenko G. W., Mazzola E. P. and Brumley W. C. (1983) Isolation and identification of Cl6 and Cl8 fatty acid esters of chloropropanediol in adulterated Spanish cooking oils. Bulletin of Environmental Contamination and Toxicology 31, 625430.
P. and Tarkowski S. (Editors) (1984) Toxic Oil Syndrome. Mass Food Poisoning in Spain. Report of a WHO Meeting: Madrid 21-25 March 1983. World
Grandjean
Health Organization Regional Office for Europe, Copenhagen. Hill R. H., Jr, Todd G. D., Kilboume E. M., Cline R. E., McGraw J., Orti D. L., Bailey S. L. and Needham L. L. (1987) Gas chromatographic/mass spectrometric determination of aniline in food oils associated with the Soanish toxic-oil syndrome. Bulletin of Environmental Contamination and Toxicology 39, 51 l-515.
Kammiiller M. E., Verhaar H. J. M., Versluis C., Terlouw J. K., Brandsma L., Penninks A. H. and Seinen W. (1988) I-Phenyl-5-vinyl-2-imidazolidinethione, a proposed causative agent of Spanish toxic oil syndrome: synthesis, and identification in one of a group of case-associated samples. Food and Chemical Toxicology 26, 119127.
Rapeseed oil manufacture and TOS Kilbourne E. M., Bernert J. T., Jr, Posada de la Paz M., Hill R. H., Jr, Abaitua Borda I., Kilbourne B. W., Zack M. M. and The Toxico-Epidemiologic Study Group. (1988) Chemical correlates of pathogenicity of oils related to the toxic-oil syndrome in Spain. American Journal of Epidemiology 127, 1210-1227. Posada M., Castro M., Kilbourne E. M., Diaz de Rojas F., Abaitua I., Tabuenca J. M. and Vioque A. (1987) Toxicoil syndrome case reports associated with the ITH oil refinery in Sevilla. Food and Chemical Toxicology 25, 87-90. Rigau-P6rez J. G., P6rez-Alvarez L., Duefias-Castro S., Choi K., Thacker S. B., Germain J. L., Gonz/dez de Andr6s G., Cafiada Royo L. and P6rez Gallardo F. (1984)
803
Epidemiologic investigation of an oil-associated pneumonic paralytic eosinophilic syndrome in Spain. American Journal of Epidemiology 119, 250-260. Tabuenca J. M. (1981) Toxic allergic syndrome caused by ingestion of rapeseed oil denatured with aniline. Lancet ii, 567-568. V~zquez Roncero A., Maestro Duran R. and Ruiz Guti6rrez V. (1984) New aniline derivatives in "toxic oil syndrome": toxicity in mice of 3-phenylamino-l,2propanediol and its fatty acid mono- and diesters. Grasas y Aceites 35, 330-331. Vioque E. and Vioque A. (1983) Presence of peroxidized fatty anilides in oils purportedly provoking toxic syndrome. Grasas y Aceites 34, 159-161.