Risk assessment of buckwheat flour contaminated by ...

This article was downloaded by: [SPLOSNA BOLNISNICA MARIBOR], [Lucija Perharič] On: 02 January 2013, At: 00:24 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Food Additives & Contaminants: Part A Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tfac20

Risk assessment of buckwheat flour contaminated by thorn-apple (Datura stramonium L.) alkaloids: a case study from Slovenia a

b

a

Lucija Perharič , Gordana Koželj , Branko Družina & Lovro Stanovnik

c

a

National Institute of Public Health, Centre for Infectious Diseases and Environmental Health, Trubarjeva 2, Ljubljana, Slovenia b

Faculty of Medicine, Institute of Forensic Medicine, Korytkova 2, Ljubljana, Slovenia

c

Faculty of Medicine, Institute of Pharmacology and Experimental Toxicology, Korytkova 2, Ljubljana, Slovenia Accepted author version posted online: 24 Oct 2012.Version of record first published: 17 Dec 2012.

To cite this article: Lucija Perharič , Gordana Koželj , Branko Družina & Lovro Stanovnik (2012): Risk assessment of buckwheat flour contaminated by thorn-apple (Datura stramonium L.) alkaloids: a case study from Slovenia, Food Additives & Contaminants: Part A, DOI:10.1080/19440049.2012.743189 To link to this article: http://dx.doi.org/10.1080/19440049.2012.743189

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Food Additives & Contaminants: Part A, 2012 http://dx.doi.org/10.1080/19440049.2012.743189

Risk assessment of buckwheat flour contaminated by thorn-apple (Datura stramonium L.) alkaloids: a case study from Slovenia Lucija Perharicˇa*, Gordana Kozˇeljb, Branko Druzˇinaa and Lovro Stanovnikc

Downloaded by [SPLOSNA BOLNISNICA MARIBOR], [Lucija Perhari] at 00:24 02 January 2013

a

National Institute of Public Health, Centre for Infectious Diseases and Environmental Health, Trubarjeva 2, Ljubljana, Slovenia; bFaculty of Medicine, Institute of Forensic Medicine, Korytkova 2, Ljubljana, Slovenia; cFaculty of Medicine, Institute of Pharmacology and Experimental Toxicology, Korytkova 2, Ljubljana, Slovenia (Received 11 April 2012; final version received 18 October 2012) In Slovenia, a mass poisoning incident involving 73 consumers with symptoms such as dry mouth, hot red skin, blurred vision, tachycardia, urinary retention, ataxia, speech disturbance, disorientation and visual hallucinations occurred in 2003. In all cases, consumers had eaten buckwheat flour food products within the last few hours. Investigations by responsible authorities identified the contamination of a range of buckwheat food products with thorn-apple (Datura stramonium L.) seeds containing toxic alkaloids, atropine and scopolamine. To ensure the safe consumption of buckwheat food products, we carried out risk characterisation and proposed provisional maximum residue levels (MRLs) of atropine and scopolamine mixture in buckwheat flour. In the absence of critical ‘‘no observed adverse effect levels’’ for atropine and scopolamine, we based our estimation of the acute reference doses on the lowest recommended therapeutic doses. Taking into account the additive effect of the two alkaloids, we calculated acute reference doses of the mixture, that is 0.05 mg/kg of body mass for atropine and 0.03 mg/kg of body mass for scopolamine. MRLs for atropine and scopolamine mixture in buckwheat flour were estimated in a worst-case scenario, that is consumption of 100 g of flour by a child weighing 10 kg and taking into account a range of atropine/scopolamine ratio in implicated food products, that is 0.85–3.3. We proposed the national MRLs for atropine/scopolamine mixture in buckwheat food products: 4.0 mg/kg (atropine) and 2.0 mg/kg(scopolamine). However, in view of the large variability in the alkaloid content, depending on the origin of the Datura, we propose that risk assessment should be carried out on a case-by-case basis, taking into account the ratio between atropine and scopolamine content in a particular sample. Keywords: atropine/scopolamine mixture; buckwheat; Datura stramonium L.; risk assessment

Introduction In Slovenia, buckwheat (Fagopyrum sp., Polygonaceae) has been very popular since the middle ages. Many traditional dishes are prepared from buckwheat grain and flour. Buckwheat is rich in vitamins, minerals and polyphenols; hence, its popularity has been increasing. Being gluten free, it is commonly used by patients suffering from gluten enteropathy (Bonafaccia et al. 2003). The amount of buckwheat harvested in Slovenia has been increasing from 560 t annually in 2003 to 1245 t in 2011 (Statistical Office of the Republic of Slovenia [cited 2012]). The precise consumption data are not available, but it is estimated that the average annual consumption of buckwheat is approximately 1 kg/person/year (Agricultural Institute of Slovenia 2012), that is 0.03 g/kg of body mass (bm)/day in a 60-kg adult. The average consumption of buckwheat in the European Union has been estimated to be 0.02 g/kg bm/day (Hjorth et al. 2010), which is 30% lower than the estimates for Slovenia, whereas the average consumption in Japan as reported by *Corresponding author. Email: [email protected] ß 2012 Taylor & Francis

Suzuki (2003) is similar to that in Slovenia. For the world’s biggest producers of buckwheat such as China, Russia and Ukraine, which in 2009 produced 570,000, 564,040 and 188,600 t, respectively (Food and Agriculture Organization of the United Nations [cited 2012]), we found no consumption data. In Slovenia, buckwheat is often marketed as ecologically grown because plant protection products are not used by buckwheat growers, which, however, may result in the presence of weeds and parasites. These may enter the food chain during harvest. Whole buckwheat grain is grey-brown, pyramidally shaped and 3–4 mm in size (Figure 1). Plants from the genus Datura (Solanaceae) are indigenous to Europe, North Africa, Asia and North and South America. The plants are upright, generally 0.5–1.5 m in height, with trumpet-shaped flowers. Seedpods or capsules may be up to 60 mm in length and are covered with 5–30 mm spines (Figure 2). The seeds are dark brown, kidney shaped and approximately 2–3 mm in length. The Datura seeds are


2

L. Perharicˇ et al.

Figure 1. Whole buckwheat grain with the seeds of thorn-apple (Datura stramonium L.) marked with the arrow.

Figure 2. Ripe thorn-apple (Datura stramonium L.) seed capsule.

relatively similar to buckwheat grain and may be difficult to distinguish and separate (Figure 1). Seedpods may yield up to 200 seeds. When seedpods dry and open, the seeds are dispersed to the soil contiguous to the plant and also disseminated through the wind and bird faeces. All parts of the plant contain atropine and scopolamine, the content of which vary depending on the geographical region location, climate, rainfall, season of the year and part of the plant (Krenzelok 2010). Literature reports on intentional and unintentional poisoning with Datura sp. date back to antiquity. In certain cultures, Datura has been used in shamanistic practices and divinatory and witchcraft rituals. It has long been known to give rise to hallucinations and illusions of sight, hearing and taste. Its repeated use, however, is believed to lead to insanity (Harner 1973). The scientific literature contains numerous reports of poisoning most commonly by ingestion or inhalation, the majority of which are intentional in order to induce the psychoactive effects, in particular the

hallucinations and vivid dreams. Ocular toxicity characterised by unilateral or bilateral mydriasis and cycloplegia, also known as ‘‘gardeners’ eye’’ and ‘‘corn pickers’ eye,’’ is a recognised consequence of the Datura sp. sap or dust entering the eye(s) (Krenzelok 2010). Literature reports of unintentional poisoning secondary to food contamination are rare (Marciniak & Sikorski 1971; Centers for Disease Control and Prevention 1984; Rwiza 1991; Smith et al. 1991; Ramirez et al. 1999). In Slovenia, the first description of mass poisoning due to the accidental contamination of lentils with Datura seeds dates to the seventeenth century (Rupel & Reisp 1969). In September 2003, the Health Inspectorate of the Republic of Slovenia was alerted by reports of domestic food poisoning after the consumption of buckwheat dishes with symptoms of classic anticholinergic syndrome. The initial analyses confirmed the presence of atropine and scopolamine in buckwheat flour. Based on the Council Regulation laying down Community procedures for contaminants in food (European Economic Community 1993), all food products containing buckwheat were temporary recalled. It was recommended that buckwheat grain should be free of Datura alkaloids ascertained by qualitative chemical analyses. However, according to buckwheat growers and importers, achieving zero level may be very difficult, if not impossible, as Datura commonly grows at the edges of buckwheat fields. To ensure the safe consumption of buckwheat food products, we carried out risk characterisation and proposed national provisional maximum residue levels (MRLs) of atropine and scopolamine mixture in buckwheat flour.

Materials and methods Risk assessment The hazard assessment, namely the hazard identification, the selection of a critical effect, the dose-response analysis, the ‘‘no observed adverse effect level’’ (NOAEL) selection, the acute reference dose (ARfD) estimation, the exposure assessment and the risk characterisation procedures, was carried out according to the international guidance (International Programme on Chemical Safety 1999; European Commission 2001; European Chemicals Bureau 2003). The recommendations of the British Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (2002) were used for the risk assessment of the alkaloid mixture.

Intake and effect inquiry The National Institute of Public Health established an ad hoc telephone self-reporting scheme. To refine the


Food Additives & Contaminants: Part A information on effects experienced by consumers, to collect more accurate exposure data and to obtain the relevant data required for the assessment of the strength of association between the exposure and the effects, we designed a questionnaire. The questionnaire was sent to consumers who had rung our telephone information line. The questionnaire included questions on age; sex; body mass and height; the type, the serial number, the source and the amount of the consumed buckwheat foods; the temporal association between the symptoms and the consumption of buckwheat foods; the course of the symptoms; the medical history; the intake of medications; and the eventual co-exposure to other chemicals at work or at home. The questionnaire was sent to the poisoned consumers retrospectively, that is 2 months after the incident. The strength of the association between the exposure and the effects was judged according to an algorithm used in pharmacoepidemiology (Strom 1989).

Buckwheat food products’ sampling Health Inspectorate of the Republic of Slovenia collected 75 samples of buckwheat grain and buckwheat food products from millers, importers, distributers and food shops throughout Slovenia: 12 wholegrain samples, that is prior to the mechanical separation of the hull, 13 samples of groats (kasha), 34 samples of flour, 8 samples of pasta, 4 samples of bread and 4 samples of semi-prepared meal zˇganci. Consumers supplied 9 samples of buckwheat foods that were believed to have caused the poisoning.

Analytical methods Buckwheat grain was examined macroscopically (Figure 1). The qualitative analyses of buckwheat flour food products were done by using gas chromatography mass spectrometry. Because of the uneven distribution of alkaloids in the flour, a larger sample of flour was required for identification. Thus, a 30-g flour sample was extracted with a fivefold volume of methanol in an Erlenmeyer flask with a ground glass joint by shaking for 15 min. Flour particles were removed by filtration, and the methanol was evaporated with nitrogen. The evaporated residue was dissolved in 10 mL of 0.05 M HCl solution and alkalinised by adding 1 mL of concentrated ammonia solution. The alkaline solution was extracted with chloroform in a separation funnel. Alkaloid containing separated organic solvent was evaporated with nitrogen, and the residue was dissolved in methanol for gas chromatography mass spectrometry analysis. Gas chromatograph HP 6890 and mass selective detector 5973 from Hewlett Packard (Palo Alto, CA, USA) were used for qualitative analysis. The compounds were separated on 30 m HP-

3

5MS ((5%-phenyl)-methylpolysiloxane)-fused silica capillary column (0.25 mm inner diameter, 0.25 mm film thickness). The temperature programme was as follows: 60 C (2 min) and then ramped to 20 C/min to a final temperature of 300 C (20 min). Temperatures of the injector, the transfer line, the ion source and the quadrupole were 275 C, 280 C, 230 C and 150 C, respectively. The flow rate of helium was 1 mL/min. One microlitre of sample was injected in the splitless mode. The obtained mass spectra were compared with the spectra of standard solutions of both alkaloids in methanol, as well as with electronic mass spectra of Pfleger-Maurer-Weber library of Mass Spectral and GC Data of Drugs, Poisons, Pesticides, Pollutants and their metabolites (Pmw_TOX3). The additional confirmation came from the retention time of the compounds of interest. Four compounds were identified in the samples of contaminated flour, namely atropine, scopolamine and their thermally degraded dehydrated products (atropine-H2O and scopolamine-H2O). A quantitative analytical method for the determination of atropine and scopolamine in buckwheat food products using liquid chromatography – tandem mass spectrometry was developed (Ku¨cˇan et al. 2004), using a liquid chromatograph (ProStar 210; Varian, Palo Alto, CA, USA) and a tandem mass spectrometer (Varian 1200 L, Palo Alto, CA, USA). A 2-g food sample was homogenised and extracted with a mixture of 20 mL of dichloromethane (high-performance liquid chromatography [HPLC] gradient, J.T. Baker, Europe, Griesheim, Germany), methanol (HPLC gradient, J.T. Baker, Europe, Griesheim, Germany) and ammonium hydroxide (Riedl de Haen, Seelze, Germany) in the ratio 70:25:5 dissolved in 18 mL of 0.5 M sulphuric acid (Kemika, Zagreb, Croatia) and alkalinised by adding 5 mL of ammonium hydroxide. The extraction was repeated twice with dichloromethane. The evaporated extract was reconstituted twice in 2 mL of methanol and filtered. The compounds were separated on a LUNA C18(2) column with a mobile phase A of 20 mM ammonium acetate HPLC gradient (Sigma Aldrich, St. Louis, MO, USA) and a mobile phase B of acetonitrile HPLC gradient (J.T. Baker, Europe, Griesheim, Germany) with a constant flow of 0.25 mL/min and gradient phase A 100% for 0.2 min, changed to 50% up to 2 min, held up to 3.2 min, changed to 100% up to 4.2 min and held unchanged up to 12 min. Argon was used as a collision gas. A mass spectrometer was used in the positive electrospray ionisation mode; transition 304 4 156, 138 was used for scopolamine and 290 4 124, 93 for atropine. Commercially available atropine and scopolamine (i.e. scopolamine hydrochloride) purchased from Sigma-Aldrich (St. Louis, MO, USA) were used as standards. The calibration coefficient was 0.998 for atropine and 0.995 for scopolamine. The signal to noise (S/N) ratio was more than 100 for fragments of

4


Table 1. Literature data (Hardman et al. 2001) on dose-dependent effects of atropine ingestion.

Ingested dose (mg) 0.5 1.0 2.0


5.0 10.0

Calculated dose assuming 60 kg body mass (bm) (mg/kg bm)

Effect Transient bradycardia, dry mouth, inhibition of sweating Dry mouth, thirst, tachycardia, moderately dilated pupils Tachycardia, severely dry mouth, severely dilated pupils, disturbed visual accommodation As above plus disturbed speech, dysphagia, restlessness, fatigue, dry and hot skin, urinary retention, reduced gut motility As above, but with more pronounced symptoms, plus pupil practically non-reactive, weak pulse, red skin, ataxia, agitation, excitation, hallucinations, delirium, coma

atropine and scopolamine. The initial level of detection for each alkaloid was 10 mg/kg, whereas the level of quantification was 30 mg/kg. These were subsequently reduced to 1 mg/kg (level of detection) and 3 mg/kg (level of quantification) by the reduction of initially uncommonly high S/N ratio to the value still higher than 10.

Results Hazard assessment Atropine (a mixture of D- in L-hyosciamine) and scopolamine (L-hyoscin) are tertiary ammonium amines with pKa values of 9.7 and 7.6–7.8, respectively. At physiological pH, tertiary amines are ionised but they are still sufficiently lipid soluble to be well absorbed through the lipid membranes, widely distributed in tissues and can enter the central nervous system (CNS). They cause competitive inhibition of acetylcholine receptors in the CNS, the vegetative ganglia and the postganglionic parasympathetic nerves, producing antimuscarinic effects also known as anticholinergic toxidrome. Peripheral effects include dry mouth and mucous membranes, mydriasis and blurred vision and tachycardia; centrally, atropine acts as a CNS stimulant, resulting in excitation, agitation, ataxia, clonus, delirium, hallucinations, seizures and coma while scopolamine acts as a CNS depressant with sleep induction, vivid dreams, amnesia and coma (Ellenhorn & Barceloux 1988; Hardman et al. 2001). Both atropine and scopolamine have been used for a long time as medicines, but they were not assessed for their safety according to modern testing protocols.

Atropine Atropine is used in anaesthesiology, cardiology and ophthalmology and as an antidote in organophosphate poisoning, while in the past it has also been used for the alleviation of symptoms in asthma (Hardman et al. 2001). In humans, atropine is quickly absorbed from

8 17 33 83 167

the gastrointestinal tract. It reaches its peak concentration 1 h after ingestion. In plasma, approximately 50% of the atropine is protein bound. It is quickly distributed throughout the body. It crosses the blood-brain barrier and the placenta; traces can be found in maternal milk. Its pharmacological effects appear once the concentration in plasma reaches 2–3 mg/L. In the liver, atropine is metabolised to tropine and tropic acid. About 60% is excreted unchanged in the urine. Its half-life in humans amounts to 2–4 h (Ellenhorn & Barceloux 1988; Hardman et al. 2001). The effects of atropine ingestion at various dose levels are shown in Table 1. The lowest reported fatal dose in adults was 32 mg given intramuscularly, while survival has been reported after ingestion of 1000 mg. The lowest reported fatal dose in a child aged 2 years and 9 months following ocular application was 1.6 mg. Factors increasing the susceptibility to atropine include extremes of age (children being particularly susceptible), high air humidity, hot climate, blond hair, chronic obstructive airway disease, coronary artery disease, Down’s syndrome, enlarged prostate, glaucoma, hiatus hernia, myasthenia gravis, paralytic ileus, pyloric stenosis, renal failure, spastic paralysis, tachycardia, such as in thyrotoxicosis, cardiac failure and post-cardiac surgery, ulcerative colitis and co-exposure to anticholinergics such as amantadine, certain antihistamines, butyrophenones, phenothiazines and tricyclic antidepressants (Ellenhorn & Barceloux 1988). The bioavailability of atropine and the doseresponse relationship are well defined for the therapeutic dose. Typical antimuscarinic effects, namely reduced saliva and sweat secretion, mydriasis and tachycardia, were noted at 3 mg/kg bm intravenously. At doses below this, bradycardia and increased saliva secretion were noted. The mechanism of the latter two effects is not entirely clear but may involve central stimulation of the vagal nuclei, partial agonism at the level of muscarinic receptors or peripheral secretion of acetylcholine by negative feedback action at the level of the M1 receptors (Lo¨nnerholm & Widerlo¨w 1975; Mirakhur 1978; Wellstein & Pitschner 1988).


Food Additives & Contaminants: Part A Animal data show a large variability in susceptibility to atropine while the symptoms in mammals are similar: dry mouth, difficulty swallowing, constipation, mydriasis, tachycardia, tachypnea, restlessness, delirium, ataxia, shaking, convulsion, respiratory depression and failure. Oral LD50 (median lethal dose) in rats amounts to 500 mg/kg, intravenous. LD50 in rats amounts to 73 mg/kg, oral LD50 min in mice amounts to 75 mg/kg, and intravenous LD50 in mice amounts to 30 mg/kg (Anonymous [cited 2003]). Effects of atropine sulphate, a form used as medicine, were observed even at the lowest dose levels used in animal studies, that is in horses at 0.01 mg/kg bm; therefore, NOAEL has not been determined. Atropine sulphate is not mutagenic. Data on long-term toxicity including carcinogenicity and reproductive toxicity are scarce (Committee for Veterinary Medicinal Products 1998).

Scopolamine Scopolamine is used to alleviate abdominal cramps, to prevent travel sickness and in anaesthesia. Therapeutic doses of 0.4–1 mg may cause fatigue, drowsiness, amnesia and non–rapid eye movement sleep in adults, as well as excitation, restlessness, hallucinations, delirium, mydriasis and tachycardia (Hardmann et al. 2001). Following ingestion it is well absorbed (Mirakhur 1978). Scopolamine is metabolised to aposcopolamine, 6-hydroxy-hyosciamine, a corresponding glucuronide and tropic acid. It crosses the blood-brain barrier and the placenta and is distributed into milk. Approximately 1% of the ingested dose is excreted in the urine. Scopolamine is more potent than atropine concerning effects on the pupillary muscles and salivary, sweat and bronchial glands while less potent concerning effects on the heart, intestinal and bronchial smooth muscles. It is sedative to the CNS. Based on experiments in dogs, an oral NOAEL has been determined at 1 mg/kg bm for a veterinary drug butyl-scopolamine. No mutagenic effects have been seen in vitro. Two-year oral studies in rats and mice have not revealed any carcinogenic potential. No teratogenic effect was seen following parenteral application to rats and rabbits, while embrytoxic effects were seen at large doses in rabbits (Anonymous [mod 2009]). A 76-year-old man suffered dyspnoea and weakness following the ingestion of 435 mg of scopolamine in home-made wine made from Datura suaveolens (Smith et al. 1991).

ARfD estimation ARfD of a chemical is an estimate of a substance in food or drinking water, expressed on body mass basis, that can be ingested over a short period of time, usually

5

during one meal or 1 day, without appreciable health risk to the consumer on the basis of all known facts at the time of evaluation. Typically, an ARfD is based primarily on an NOAEL (European Commission 2001). In the absence of critical NOAELs for atropine and scopolamine, we based our estimation of ARfDs on the lowest recommended paediatric and obstetric therapeutic doses, that is 10 mg/kg bm for atropine (Mirakhur & Jones 1982; Honkavaara & Pyykko¨ 1999) and 5 mg/kg bm for scopolamine (Pihlajama¨ki et al. 1986). We applied an uncertainty factor of 100, 10 in view of using therapeutic doses rather than NOAELs and 10 to account for the large variability in susceptibility for tropane alkaloids, leading to ARfDs of 0.1 mg/kg bm and 0.05 mg/kg bm for atropine and scopolamine, respectively. In view of the fact that Datura contains both alkaloids and that both act as competitive inhibitors of muscarinic receptors (Hardman et al. 2001), we took into account the additive effect of the mixture (Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment 2002), giving the ARfD of the mixture as 0.05 mg/kg bm for atropine and 0.03 mg/kg bm for scopolamine.

Exposure assessment The initial exposure assessment was based on a recipe for the preparation of buckwheat zˇganci (Kalinsˇ ek 1956), a buckwheat flour dish that was consumed in the majority of identified poisoning cases. According to this recipe, 100 g of flour is boiled in salty water, which makes a portion for an adult. Zˇganci may be eaten with warm milk as breakfast cereal or as a main or side lunch or dinner dish usually larded with cracklings. Examination of whole buckwheat grain identified 7–190 seeds of thorn-apple (Datura stramonium L.) per kilogram of grain, while no Datura seeds were found in buckwheat groats. The qualitative analyses of nine buckwheat flour food samples confirmed the presence of atropine and scopolamine in all nine samples supplied by the poisoned consumers. In 18 buckwheat flour food samples, the levels of atropine and scopolamine were above the initial level of detection of 10 mg/kg. The levels of atropine ranged from 30 to 26,000 mg/kg, and of scopolamine from 45 to 12,000 mg/kg with level of quantification of 30 mg/kg. The ratio between atropine and scopolamine ranged from 0.85 to 3.3 (Table 2).

Hazard and exposure assessment in our case series Questionnaires were sent to 73 consumers with symptoms of tropane alkaloid toxicity identified through


6


our ad hoc telephone self-reporting scheme. We, received 46 replies. The youngest subject affected was a 6-week-old baby exposed via maternal milk; the rest were aged from 10 months to 82 years. They ingested food containing buckwheat flour from 45 minutes to 8 hours prior to the appearance of the symptoms. The most commonly reported symptoms were dry mouth and throat (42), dizziness (29), visual disturbances (26), red skin (24), hot skin (22), nausea (21), confusion (21) and fast heartbeat (20). All reported symptoms are presented in Figure 3. The strength of association

between the ingestion of buckwheat flour–containing meal and reported symptoms is shown in Table 3. Temporal association was established in all 46 cases. None of the consumers suffered diseases nor were exposed to medication, which could have been responsible for the experienced symptoms. In five consumers, the symptoms reappeared after the exposure to the same food product. The symptoms disappeared spontaneously in all subjects within 48 hours. Of the 46 consumers who answered the questionnaire, the majority had eaten home prepared

Table 2. Quantitative analyses of contaminated buckwheat products.

Sample no. 1751 1752 1753 1754 1755 1762 1764 1765 1766 1767 1768 1770 1772 1774 1775 1777 1792 1796

Type of food Buckwheat flour Buckwheat flour Buckwheat flour Buckwheat flour Semi-prepared buckwheat ‘‘zˇganci’’ Buckwheat flour Buckwheat flour Buckwheat flour Buckwheat flour Buckwheat dumplings Fresh pasta with buckwheat Buckwheat pasta Semi-prepared buckwheat ‘‘zˇganci’’ Buckwheat flour Buckwheat flour Buckwheat pasta Semi-prepared buckwheat ‘‘zˇganci’’ Buckwheat flour

Origin of buckwheat Hungary The Czech Hungary Hungary Hungary Hungary China The Czech The Czech Hungary Hungary Hungary China The Czech Hungary Hungary Hungary Slovenia

Republic

Republic Republic

Republic

Atropine (mg/kg)

Scopolamine (mg/kg)

Atropine/scopolamine ratio

26,000 30 500 87 160 375 350 72 530 330 1800 3600 160 38 600 380 60 50

12,000 530 275 530 75 190 230 82 36 100 1500 3100 100 530 700 190 530 30

2.16 NA 1.81 NA 2.13 1.97 1.50 0.88 NA 3.30 1.20 1.16 1.60 NA 0.85 2.00 NA 1.70

Notes: Level of detection, 10 mg/kg; level of quantification, 30 mg/kg. NA, not applicable.

Figure 3. Occurrence of symptoms experienced by the 46 consumers who answered the questionnaire.


Food Additives & Contaminants: Part A buckwheat zˇganci (40), semi-prepared (instant) buckwheat zˇganci (3), whole wheat bread in which the presence of buckwheat flour was questionable (2) or buckwheat pasta (1). Eight consumers were not able to estimate how much meal they had consumed, three reported having used approximately 50 g of flour, eighteen 100 g, two 125 g, ten 150 g, four 200 g and one 250 g. The amount of buckwheat flour used for preparation of zˇganci ranged from 100 to 250 g/adult (98th percentile was 200 g). All five children aged from 10 months (bm 9.5 kg) to 5 years have ingested zˇganci made from 50 to 100 g of buckwheat flour. On the basis of answers from the questionnaire and the laboratory analysis of the ingested food (Table 2), we were able to quantify the exposure to alkaloids in 12 cases. We estimated that the poisoned consumers ingested from 0.7 to 137.6 mg of atropine/kg bm and 0.4 to 63.5 mg of scopolamine/kg bm, respectively. Table 3. Strength of association between the ingestion of buckwheat products and symptoms based on the algorithm by Strom (1989).

Association

Number of subjects

Possible

2

Probable

39

Highly probable

5

Comment Symptoms appeared after consuming bread that ought not contain any buckwheat flour Symptoms disappeared on dechallenge; no other causes, such as co-exposure to medication or other chemicals or presence of diseases, were identified Symptoms reappeared on re-challenge

7

Table 4 shows the comparison between the lowest therapeutic doses of both alkaloids reported in the literature, the lowest toxic dose estimated in our case series and the proposed ARfDs.

Risk characterisation and MRL proposal We assumed that it would be unlikely that more than one buckwheat dish would be consumed within the same meal and that other foods within the same meal would be contaminated with Datura alkaloids. Therefore, we calculated MRLs for atropine and scopolamine mixture in buckwheat flour in a worstcase scenario, that is consumption of 100 g of flour in a child weighing 10 kg, and applying the ARfD of the mixture, that is 0.05 mg/kg bw for atropine and 0.03 mg/kg bw for scopolamine: buckwheat flour should not contain more than 5 mg of atropine/kg and 3 mg of scopolamine/kg of flour. In view of the large variability in the atropine/ scopolamine ratio between different Datura specimens in our food samples (Table 2), we also took into account the highest and the lowest ratio between atropine and scopolamine. We calculated the corresponding provisional MRLs: 6.1 mg/kg of flour for atropine and 1.9 mg/kg of flour for scopolamine (ratio 3.3), and 3.7 mg/kg of flour for atropine and 4.3 mg/kg of flour for scopolamine (ratio 0.85). Thus, to cover both extremes, we proposed provisional national MRLs of 4 mg/kg of flour for atropine and 2 mg/kg of flour for scopolamine.

Discussion Mass poisoning with buckwheat food products contaminated with D. stramonium L. seeds prompted the risk assessment for a mixture of atropine and

Table 4. Therapeutic doses, lowest toxic doses in our case series and the estimated acute reference doses (ARfD) of atropine and scopolamine.

Contaminant

The lowest therapeutic dose reported in the literature (mg/kg bm)a

The lowest toxic dose in our case series (mg/kg bm)b

ARfD – calculated in our preliminary risk assessment (mg/kg bm)c

Atropine Scopolamine

10.0 5.0

0.71 0.34

0.05 0.03

Notes: aMirakhur and Jones (1982), Pihlajama¨ki et al. (1986) and Honkavaara and Pyyko¨ (1999). b The doses were estimated from the amount of incriminated food eaten as reported by the poisoned consumers, their reported body mass (bm) and the amounts of alkaloids determined in the corresponding food samples. c Calculation was based on the lowest clinical doses found in the literature (Mirakhur & Jones 1982; Pihlajama¨ki et al. 1986; Honkavaara & Pyyko¨ 1999), applying the uncertainty factor of 100 (10 for the absence of ‘‘no observed adverse effect level’’ and 10 for the variability in sensitivity) and taking into account the assumed additive effect of atropine and scopolamine (Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment 2002).


8


scopolamine as food contaminants. To our knowledge, this was the first quantitative risk assessment of a mixture of atropine and scopolamine. However, there were a number of uncertainties in our risk assessment and estimation of MRLs. In the absence of NOAELs for each alkaloid, we derived ARfDs from the paediatric and obstetric therapeutic doses, which, therefore, represented an important uncertainty in our hazard assessment. However, the subsequent analyses of responses from the poisoned consumers showed that the lowest toxic dose of atropine in our case series was 14 times higher than our estimated ARfD, while for scopolamine it was 11 times higher than our estimated ARfD (Table 4). This finding was suggestive that the safety margin in our risk assessment was probably adequate. A further confounding factor was self-reporting of the symptoms, which might have been influenced to some degree by the extensive media coverage of the event. Another uncertainty in our risk assessment was the amount of buckwheat flour or other buckwheat food products consumed. In the initial exposure assessment, we used the information from a recipe for the preparation of zˇganci using 100 g of flour/adult portion. The retrospectively obtained information from the poisoned subjects suggested that the 98th percentile of poisoned adults consumed 200 g of buckwheat flour/ day (3.3 g/kg bm/day), while the five children aged up to 5 years consumed zˇganci made of 100 g of flour or less (10 g/kg bm/day). Yet the number of consumers in our case series was too small for a statistically valid stratification. The accuracy of the exposure was questionable to some degree due to a 2-month delay in obtaining the information on the amount of buckwheat flour food products consumed by the poisoned subjects rather than a 24-h recall as recommended by the World Health Organization (2002). At the time of our poisoning incident, we did not find any relevant (i.e. high consumption) data on buckwheat consumption in other countries. Subsequently, the World Health Organization has published 97.5th percentile consumption figures for various commodities, which included data on buckwheat consumption from the Netherlands: 1.86 g/kg bm/day in the general population and 3.56 g/kg bm/day in children 6 years or older (World Health Organization [mod 2008]). Our data indicate that the consumption of buckwheat among the Slovenian consumers is higher although our consumption data are based on a small number of consumers. To define the exposure more precisely, it would be necessary to include buckwheat food products in national nutritional surveys including various age groups and subgroups with specific dietary habits, such as gluten enteropathy sufferers, vegetarians, vegans and consumers eating predominantly organically grown food; all of these subgroups may consume more buckwheat than the general population.

The variability of the atropine/scopolamine ratio in plants represents yet another uncertainty. The amount of alkaloids is influenced by the geographical region, climate, rainfall, season of the year and part of the plant (Krenzelok 2010). As shown in Table 2, the origin of buckwheat in our incident was from four different countries and the sample was too small to allow the determination of the alkaloid ratio pattern depending on the country of origin. The lowest ratio was 0.85, and the highest was 3.3; in both cases, the buckwheat originated from Hungary. According to the European Food Safety Authority (2008), the seeds of D. stramonium from Bulgaria, Russia and the United States contained four times more atropine than scopolamine; in seeds from Italy, the content of atropine was 14 times higher, whereas in seeds from Poland, the content of atropine was 10%–30% lower. In Slovenia, the highest alkaloid content in D. stramonium was found in seeds (0.24% atropine/0.05% scopolamine), followed by leaves (0.21% atropine/0.07% scopolamine), flowers (0.11% atropine/0.07% scopolamine) and roots (0.05% atropine/0.009% scopolamine). Plant samples gathered in the Mediterranean region of the country contained up to six times more atropine than scopolamine in comparison to those gathered in central Slovenia and the Alpine region (Klancˇar et al. 2006). In view of the large variability in alkaloid content in the Datura plants, we propose that risk assessment should be done on a case-by-case basis by taking into account the ratio between atropine and scopolamine content in a particular sample. To ensure the safety of buckwheat food products, we proposed the monitoring of atropine and scopolamine in buckwheat flour food products in Slovenia. Following the introduction of monitoring, no further intoxications associated with the consumption of buckwheat food products have been identified, which is an indication that our risk assessment was probably appropriate. However, the reason for no further poisoning reports might be related to the fact that some time after the mass poisoning incident the attention of consumers has probably ceased and it may be that mild transient symptoms, even if experienced, would not have been ascribed to the consumption of buckwheat food products. A search for a hazard category biotoxins in the Rapid Alert System for Food and Feed portal from 2006 to 2011 revealed six notifications concerning the presence of Datura seeds: five in millet from Austria and Hungary, one in vegetable and bacon stir fry mix from Spain and one in canned green beans from Hungary. In 2008, there was a notification on the presence of Hyoscyamus niger (another tropane alkaloids-containing plant) seeds in blue poppy (Meconopsis betonicifolia) seeds from the Czech Republic (Rapid Alert System for Food and Feed portal [created 2006]).


Food Additives & Contaminants: Part A In the European Union, the contamination of food products with weeds containing tropane alkaloids such as thorn-apple (D. stramonium L.) is rare, though more likely in grain grown without the use of plant protection products, such as buckwheat. The population groups particularly at risk are small children, people living in hot and humid environment, patients, especially those sensitive to antimuscarinic action, and people who may consume more buckwheat food products, such as gluten enteropathy sufferers, vegetarians, vegans and those who use predominantly organically grown foods. In light of possible severe acute toxicity, caused by the consumption of contaminated buckwheat food products, we believe that continuous surveillance of this kind of foods is justified.

References Agricultural Institute of Slovenia. Assessment of situation in agriculture in 2009 [Internet]. Ljubljana (Slovenia); [cited 2012 Oct 1]. Available from: www.kis.si/datoteke/file/kis/ SLO/EKON/.../SpZP-2009-trgi.doc Anonymous. Atropine [Internet]. Bethesda (MD): U.S. National Library of Medicine; [created 1983 Apr 1; modified 2003 Sept 12; cited 2003 Oct 8]. Available from: http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/ PoDVdz:animal Anonymous. Scopolamine [Internet]. Bethesda (MD): U.S. National Library of Medicine; [created 1983 Apr 1; modified 2009 May 5; cited 2003 Oct 8]. Available from: http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./ temp/BNjCQ9:1 Bonafaccia G, Gambelli L, Fabjan N, Kreft I. 2003. Trace elements from flour and bran from common and tartary buckwheat. Food Chem. 83:1–5. Centers for Disease Control and Prevention. 1984. Datura poisoning from hamburger – Canada. MMWR Morb Mortal Wkly Rep. 33:282–283. Committee for Veterinary Medicinal Products. 1998. Atropine. London, UK: European Medicines Control Agency. Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment. 2002. Risk assessment of mixtures of pesticides and similar substances. London, UK: Food Standards Agency. Ellenhorn MJ, Barceloux DG. 1988. Medical toxicology: diagnosis and treatment of human poisoning. New York: Elsevier. Chapter 41, Plants-mycotoxins-mushrooms; p. 1214–1351. European Chemicals Bureau, Institute for Health and Consumer Protection. 2003. Technical guidance document on risk assessment. Ispra (Italy): European Commission, Joint Research Centre. European Commission, Health and Consumer Protection Directorate General. 2001. Draft guidance document. Guidance for the setting of an acute reference dose (ARfD). 7199/VI/99 rev. 5. Brussels (Belgium): European Commission.

9

European Economic Community. 1993. Council Regulation (EEC) No 315/93 of 8 February 1993 laying down Community procedures for contaminants in food. Off J Eur Com. L. 36:1–3. European Food Safety Authority. 2008. Scientific opinion of the panel on contaminants in the food chain on a request from the European Commission on tropane alkaloids (from Datura sp.) as undesirable substances in animal feed. EFSA J. 691:1–55. Food and Agriculture Organization of the United Nations. FAOSTAT. Food and Agriculture commodities production [Internet]. Rome (Italy); [cited 2012 Oct 1]. Available from: http://faostat.fao.org/site/339/default.aspx Hardman JG, Limbird LE, Goodman Gilman A. 2001. Goodman and Gilman’s The pharmacological basis of therapeutics. 10th ed. New York: McGraw Hill. Chapter 7, Muscarinic receptor agonists and antagonists; p. 162–173. Harner MJ. 1973. Hallucinogens and shamanism. New York: Oxford University Press. Chapter 8, The role of hallucinogenic plants in European witchcraft; p. 125–150. Hjorth K, Strange Hermann S, Bjerre Christensen H, Erecious Poulsen M. 2010. Cereals and feeding stuff – production, consumption and pesticides (Version 4) [Internet]. [cited 2012 Oct 1]. Lyngby (Denmark): European Union Reference Laboratory for Cereals and Feeding Stuff, National Food Institute, The Danish Technical University. Available from: http://www.crl-pesticides.eu/ library/docs/cf/Cereals%20and%20feedingstuff%202009% 20_4_.pdf. Honkavaara P, Pyykko¨ I. 1999. Effects of atropine and scopolamine on bradycardia and emetic symptoms in otoplasty. Laryngoscope. 109:108–112. International Programme on Chemical Safety. 1999. Principles for the assessment of risk to human health from exposure to chemicals: environmental health criteria 210. Geneva, Switzerland: World Health Organization. Kalinsˇ ek F. 1956. [Slovene cookery book]. 13th ed. Ljubljana, Slovenia: Slovenski knjizˇnji zavod; p. 103. Klancˇar U, Kac J, Mlinaricˇ A, Krbavcˇicˇ A. 2006. [Content of atropine and scopolamine in poisonous solanaceae plants from Slovenia]. Zdrav Vestn. 75:157–162. Krenzelok EP. 2010. Aspects of Datura poisoning and treatment. Clin Toxicol. 48:104–110. Ku¨cˇan L, Sˇtajnbaher D, Zidaricˇ M. 2004. [Determination of atropine and scopolamnie in buckwheat products by LC/MS/MS]. Proceedings of the meeting Slovenian Days of Chemistry; 2004 Sept 23–24; Maribor, Faculty of Chemistry and Chemical Technology; p. 15–22. Lo¨nnerholm G, Widerlo¨w E. 1975. Effect of intravenous atropine and methylatropine on heart rate and secretion of saliva in man. Eur J Clin Pharmacol. 8:233–240. Marciniak J, Sikorski M. 1971. Intoxication with alkaloids of Datura stramonium and Datura inoxia following honey ingestion. Pol Tyg Lek. 27:1002–1003. Mirakhur RK. 1978. Comparative study of the effects of oral and i.m. atropine and hyoscine in volunteers. Br J Anaesth. 50:591–598. Mirakhur RK, Jones CJ. 1982. Atropine and glycopyrrolate: changes in cardiac rate and rhythm in conscious and anaesthetized children. Anaesth Intensive Care. 10:328–332. Pihlajama¨ki KK, Kanto JH, Oksman-Caldenty KM. 1986. Pharmakokinetic and clinical effects of scopolamine in


10


caesarean section patients. Acta Pharmacol Toxicol (Copenh.). 59:259–262. Ramirez M, Rivera E, Ereu C. 1999. Fifteen cases of atropine poisoning after honey ingestion. Vet Hum Toxicol. 41:19–20. Rapid Alert System for Food and Feed portal [Internet]. Brussels (Belgium): DG Sanco; [created 2006 Jan; cited 2012 Mar 28]. Available from: https://webgate.ec.europa.eu/rasff-window/portal/ Rupel M, Reisp B. 1969. [Valvasor’s reading]. 2nd ed. Ljubljana, Slovenia: Mladinska knjiga. Chapter 11, [Towns, markets, castles and monasteries of Carniolia]; p. 336–338. Rwiza HT. 1991. Jimson weed food poisoning: an epidemic at Usang rural government hospital. Trop Geogr Med. 43:85–90. Smith EA, Meloan CE, Pickel JA, Oehme FW. 1991. Scopolamine poisoning from homemade ‘‘moon flower’’ wine. J Anal Toxicol. 15:216–219. Statistical Office of the Republic of Slovenia. Production of cereals in Slovenia [Internet]. Ljubljana (Slovenia); [cited 2012 Oct 1]. Available from: www.stat.si/ PrikaziDatoteko.aspx?id=4719

Strom BL. 1989. Determining causation from case reports. Pharmacoepidemiology. New York: Churchill Livingstone; p. 275–288. Suzuki I. 2003. Production and usage of buckwheat grain and flour in Japan. Fagopyrum. 20:13–16. Wellstein A, Pitschner HF. 1988. Complex dose-response curves of atropine in man explained by different functions of M1- and M2- cholinoceptors. Naunyn-Schmiederberg’s Arch Pharmacol. 338:19–27. World Health Organization. 2002. Methodology for exposure assessment of contaminants and toxins in food. Geneva: World Health Organization; p. 1–19. World Health Organization. Global Environment Monitoring System/Food for the Codex Committee on Pesticide Residues and the joint Food and Agriculture Organization/WHO meetings on pesticide residues: highest reported 97.5th percentile consumption figures (eaters only) for various commodities by the general population and children ages 6 and under [Internet]. Geneva (Switzerland); [modified 2008 Apr; cited 2011 Dec 20]. Available from: http://www.who.int/foodsafety/chem/en/ acute_hazard_db1.pdf