Environment International 49 (2012) 115–119

Contents lists available at SciVerse ScienceDirect

Environment International

j ourna l homepage: www.e lsev ie r .com/ locate /env int

Exposure to domoic acid through shellfish consumption in Belgium

M. Andjelkovic a,⁎, S. Vandevijvere a,⁎⁎, J. Van Klaveren b, H. Van Oyen a, J. Van Loco a

a Scientific Institute of Public Health, Juliette Wytsmanstraat 14, B-1050 Brussels, Belgiumb Wageningen University, Wageningen, The Netherlands

Abbreviations: NBFCS, National Belgian Food Consacid; ASP, amnesic shellfish poisoning; SPE, solid phasemance liquid chromatography; TDI, tolerable daily intak⁎ Corresponding author. Tel.: +32 2 642 5592; fax: +

⁎⁎ Corresponding author. Tel.: +32 2 642 5029; fax: +E-mail addresses: mirjana.andjelkovic@wiv-isp.be (M

stefanie.vandevijvere@wiv-isp.be (S. Vandevijvere).

0160-4120/$ – see front matter © 2012 Elsevier Ltd. Allhttp://dx.doi.org/10.1016/j.envint.2012.08.007

a b s t r a c t

a r t i c l e i n f o

Article history:Received 7 June 2012Accepted 17 August 2012Available online 23 September 2012

Keywords:Marine toxinsDomoic acidShellfishShellfish consumptionExposure assessment

A main known culprit causing amnesic shellfish poisoning in humans is domoic acid (DA). The toxin appear-ance in sea waters (by counting the toxin producing algae) and consequently in shellfish is closely monitoredto prevent acute intoxications with gastrointestinal symptoms and neurological signs. However it is assumedthat there might be some chronic problems with repetitive exposures to the toxin in animals. In humans thisis greatly unknown and it is mostly assessed by relating reported toxin episodes and representative con-sumption data. Although in Belgium no alarming outbreaks have been reported in recent years, different con-centrations of DA have been found in shellfish samples. In this study the human acute and chronic exposureto DA through shellfish consumption was evaluated by linking the data of DA concentrations in samplescollected in the scope of the National Food control program in the period 2004–2009 and consumptiondata obtained from the National Belgian Food Consumption Survey including 3245 adults. The found levelof toxin was highest in scallops while lowest in mussels. The mean usual long-term intake of molluscssuch as scallops, mussels and oysters for the whole Belgian population was from 0.10 g/day for scallops to1.21 g/day for mussels. With average portion size estimated to be 56–108 g/day depending on the shellfishsource it was calculated that less than 1% of the population would be at risk of acute intoxication. Using a me-dium bound approach, 5–6% of the population shows chronic exposure exceeding the tolerable daily intake of0.075 μg/kg bw per day with scallops being the most probable toxin vector when using lower (68.5%) andmedium (45.6%) bound concentrations.

© 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Domoic acid (DA) is a marine biotoxin causing amnesic shellfishpoisoning (ASP) in humans. Symptoms of ASP include gastrointestinaleffects (nausea, vomiting, diarrhoea or abdominal cramps) and/or neu-rological signs (confusion, loss of memory, or other serious signs suchas seizure or coma) occurring within 24 to 48 h after ingestion, respec-tively. This toxin was first detected by monitoring services in Canadaand first identified in 1987 (Bates et al., 2012), in the United States,and in a number of European countries (EFSA, 2009). The NationalFood Control Program in Belgium detects this toxin in the samplescollected each year from either the harvest zone or retail or fish market.Toxic blooms of DA-producing diatoms are a global and emerging issue(Trainer et al., 2009) and appear to be increasing in frequency and toxic-ity, thereby presenting a continued threat to human health and seafoodsafety (Lefebvre and Robertson, 2010). While the evaluation of the DA

umption Survey; DA, domoicextraction; HPLC, high perfor-e.32 2 642 5736.32 2 642 5410.. Andjelkovic),

rights reserved.

concentration is performed in many countries on a regular basis, the ex-posure assessment of consumers to DA,which relies both on reported oc-currences and representative consumption data, is scarce. The limitedconsumption data pose a considerable breach in evaluating the exposurerisk of toxins present in shellfish products.

DA is produced by diatoms Pseudo-nitzschia, mostly (Lefebvre andRobertson, 2010) and by marine red algae of the genus Chondria(EFSA, 2009) and has been reported to accumulate in a wide varietyof seafood, including mussels, scallops and anchovies. Of those, mus-sels have relatively rapid uptake and depuration rates while smallersized oysters would retain more DA and eliminate it slower than biggersized oysters (Mafra et al., 2010). Scallops accumulate the toxin for a lon-ger time (Blanco et al., 2002). This may be explained by the lack of cellmembrane transporters to eliminate free DA from the cytosol (Maurizand Blanco, 2010).

Scallops were reported to be the most affected species in Irelandwaters (James et al., 2005), where during a 10 year period around11% of controlled shellfish samples were detected to contain morethan 20 mg/kg DA.

France reported DA in 8% of the controlled samples, while the con-trol program in the United Kingdom detected DA in 17% of all testedshellfish samples with two very large outbreaks in 1999 and 2002(Hinder et al., 2011). Some areas were shown to be less affected likeSpain and Portugal.

116 M. Andjelkovic et al. / Environment International 49 (2012) 115–119

From non-bivalve species, crabs may accumulate DA. Howevertheir ability to accumulate and eliminate the toxin is different frombivalves. Whereas in the latter DA is excreted, in crabs it is rapidlyand extensively absorbed from the digestive system (Schultz et al.,2008). On the other hand there is no evidence that DA is toxic tofish although they can accumulate (e.g. salmon) it up to levels of1800 μg/g bodyweight. The fish is depurated within few days butmay be a vector of toxin to higher fish eating animals during thisperiod.

Enforcement in the EU of the regulatory limit for DA of 20 μg/kg(Regulation EC, 2074/2005) implies that samples with DA levelsbelow this value would be released on the market and get into thefood chain (Bates et al., 2012; EC, 2074/2005). Consumption of theseproducts may not result in an acute intoxication with DA but it is notyet clear what would be the effect of their long-term consumption.Furthermore, some population groups like children and infants are athigher risk due to their lower bodyweight. In addition in rats transitionof the toxin viamilk to neonateswas proven to be also a route (Maucherand Ramsdell, 2005).

Risks to human health via low level exposure are not yet fully un-derstood, as there are only studies on animals or model systems. Thetoxicological studies with animals show clinical signs and brain le-sions mostly consistent with those in naturally exposed humans andsea lions. Those clinical signs included seizures, periods of marked leth-argy and inappetence, vomiting, muscular twitching, central blindnessblepharospasm (often unilateral) and abnormal behaviour (Goldsteinet al., 2008). Based on the data for sea lions the authors suggested thatthe conversion from acute to chronic exposure is possible. In a modelsystem with rats, seizures with increasing magnitude over a period of6 months were induced in animals by application of low repetitivelyDA doses (Muha and Ramsdell, 2011).

Low levels of DA (0.20–0.75 ppm) show no toxic symptoms inhumans and non-human primates, but clinical effects are apparentin them and in humans, at a concentration of 1.0 ppm. The tolerabledaily intake (TDI) of DA for humans is calculated as 0.075 ppm,whereasfor razor clams and crabs, the TDIs are 19.4 and 31.5 ppm respectively(Costa et al., 2010; Kumar et al., 2009).

The marine mammals (sea lions, sea otters and dolphins) were themarine species most often affected by DA in recent years. Moreover,the rate of intoxication events due to DA appears to be increasing(Lefebvre and Robertson, 2010). The toxin is transferred by dietaryconsumption among the species. Humans consume many of thesemarine species. Although no alarming outbreaks have been reportedin recent years, DA has been found in different concentrations in shell-fish samples in Belgium. In this study the human acute and chronic ex-posure to DA through shellfish consumption was evaluated by linkingthe data of DA concentrations collected in the scope of the FederalAgency for the Safety of the Food Chain control program in the period2004–2009 and consumption data obtained from the National BelgianFood Consumption Survey (NBFCS) in adults (2004).

2. Materials and methods

2.1. Shellfish samples

During the period 2004–2009 the Federal Agency for the Safety ofthe Food Chain collected 224 mussel samples, 217 oyster samples,319 scallop samples and 45 shellfish samples of unspecified origin,indicated hereafter as molluscs. The samples originated from the Na-tional control program and were analysed by the Scientific Institute ofPublic Health for the presence of domoic acid. The samples were de-livered fresh and analysed within a maximum 4 days upon arrival.The cleaning and initial preparation were done according to the pro-cedure described by the European Reference Laboratory for marinetoxins (EURL-MB, 2008).

2.2. Reagents and solutions

Solvents used for the HPLC analysis were methanol and acetonitrileof HPLC grade supplied by Merck and Millipore-Q cleaned water. Certi-fied DA standard was purchased from the National Research Council ofCanada (NRC) for periodic calibration curve.

The DA standard (SIGMA-Aldrich, Bornem, BE) was used in theroutine analyses for preparing the standard solution used for spikingthe blank sample (one part) and preparing the control sample.

2.3. Toxin extraction

Extractions were carried out according to the method of Quilliam(1995) with some modifications. After homogenizing 4 g of a samplethe extractionwas performedwith aqueous 50%methanol (4:1)mixingin a vortex for 3 min. After 10 min of centrifugation at 5000 rpm, thesupernatant was purified through SPE-SAX (SupelClean, 3 ml, Supelco,Bornem — Belgium) columns filtered into a screw-cap autosamplervial with a filter (0.45 mm, Milipore millex — HV). In order to controlthe recovery, the blank sample and the control samples were analysedin each injection series. The latter was spiked with a standard solutionto a final concentration of 25 μg/g.

2.4. Quantification of domoic acid

Liquid chromatographywas performed on aVarianHPLC-UV system(Model 310) equipped with a binary pump (Varian, model 230) and anautosampler (Varian,model 410). Data collectionwas performed by theVarian Star Chromatographic Workstation. The column used was a C18Bondapak HPLC column (Waters 300×3.9 mm, 10 μm) with isocraticgradient of the mobile phase containing 87% aqueous solution of 8.5%phosphoric acid adjusted to pH 2.5 and 13% of 100% acetonitril. This so-lution was prepared freshly before each analysis. The injection volumewas 20 μl and analysis time was set at 20 min. Detection wavelengthwas set at 242 nm.

Calibration was performed by a full set of calibration standards ofDA (1–25 mg/ml). Calibration curves obtained with a double pointcalibration were always linear, with a correlation coefficient of 0.99.Under these conditions the detection limit was 0.8 μg/g tissue.

2.5. Food consumption data

Consumption data from the National Food Consumption survey2004 were used to perform the exposure assessments. Aims, designand methods of this survey are described elsewhere (De Vriese et al.,2005; Vandevijvere et al., 2009). The target population comprised allBelgian inhabitants of 15 years or older.

The sample included 3245 participants randomly selected fromthe National Register, using a multi-stage stratified procedure. Informa-tion on dietary intake was collected by two non-consecutive 24-hour re-calls in combination with a food frequency questionnaire. The 24-hourrecall was carried out using EPIC-SOFT software (Slimani et al., 2002).This program allows obtaining very detailed information about thefoods consumed and the recipes used in a standardized way.

3083 participants (1537 women and 1546 men) completed two24-hour recalls. Participants were categorized into four age groups:15–18 years (n=760), 19–59 years (n=830), 60–74 years (n=789)and 75 years or older (n=704).

2.6. Statistical analysis

2.6.1. Acute exposure to domoic acidThe acute dietary exposure to domoic acidwas calculated via a prob-

abilistic approach using the “Monte Carlo risk assessment program(MCRA)”, Release 6.2, available for registered users at the RIVMwebsite(de Boer and Van der Voet, 2007). For the estimation of daily acute

Table 2Mean consumption of shellfish in Belgium (National Food Consumption survey, 2004).

Food Mean consumption(g)

Mean, consumptiondays only(g)

Number ofconsumptiondays

Totalnumberof days

All days p95 Consumptiondays only

p95

Molluscsn.s.

0.04 0 250 250 1 6488

Mussels 1.21 0 107.93 250 73 6488Oysters 0.14 0 103.24 252 9 6488Scallops 0.10 0 56.45 149 12 6488

n.s., non specified.

Table 3Percentiles of intake of domoic acid (acute exposure assessment), medium boundapproach, consumption days only, including uncertainty of percentile distribution.

117M. Andjelkovic et al. / Environment International 49 (2012) 115–119

exposure, daily consumption patterns selected from the consumptiondatabase, were multiplied with randomly selected domoic acid levelsfrom the concentration database for each food group (mussels, scallops,oysters and shellfish of undetermined origin) and expressed per kgbodyweight. Because a person eating shellfishwill not eat the samepor-tion size containing the same level of toxin each time, the probabilisticcalculation includes all combinations of the different occurrenceand consumption data. To determine the exposures at each percen-tile, 100,000 iterations were performed for the general population.

2.6.2. Chronic exposure to domoic acidSimilarly like above the long-term dietary exposure to domoic acid

was calculated via a probabilistic approach using MCRA. All daily con-sumption patterns were multiplied with the mean domoic acid concen-tration value at the food group level (mussels, scallops, oysters andshellfish of undetermined origin), and summed over foods consumedper day and individual. The estimated exposures were adjusted for theindividual's body weight. A distribution of daily exposures, calculatedas described above using mean concentrations of domoic acid per foodor food group, includes both the variation between individuals and be-tween the days of one individual. However, to assess the long-term in-take within a population only the former type of variation is of interest,since in the long run the variation between different days of one individ-ual will level out.

Therefore, to calculate a long-term dietary exposure distribution,the within-person (between days) variation should first be removedfrom the distribution of daily exposures using statistical models. Therelatively new beta-binomial-normal (BBN) model was used (Slob,2006).

To remove the within-person variation from the daily exposures,the BBNmodel transforms the daily exposure distribution into a normaldistribution using a logarithmic function. After removal of the within-person variation, the normal distribution is back-transformed and isthen considered a long-termdietary exposure distribution. Transforma-tion of the daily exposure distribution into a normal distribution is animportant prerequisite to be able to use statistical models that assessthe long-term exposure. To determine the exposures at each percentile,1,000,000 iterations were performed for the general population.

For both chronic and acute exposure lower,mediumandupper boundscenarios were calculated. In the lower bound scenario all non detectswere set to 0 mg/kg, while in the medium and upper bound scenariothe non detects were set at half of the limit of quantification (LOQ=0.5 mg/kg) and LOQ (=1 mg/kg), respectively.

3. Results

3.1. Domoic acid concentration in shellfish samples

The domoic acid concentrations found in the different samples ofmolluscs, mussels, oysters, and scallops can be found in Table 1. Meanconcentration was highest in scallops whereas mean positive concen-tration (representing the concentration above the legislative norm of20 mg/kg) was highest in oysters. Both mean concentration and mean

Table 1Domoic acid concentration in collected samples (Federal Agency for the Safety of theFood Chain, 2004–2009, Belgium).

Food Meanconcentration(mg/kg)

Mean positiveconcentration(mg/kg)

Range(mg/kg)

Positivesamples

Totalsamples

Molluscs n.s. 6.5 48.75 0.80–187.70 6 45Mussels 0.06 0.82 0.53–1.67 16 224Oysters 0.35 77 77 1 217Scallops 7.14 32.54 0.80–203.40 70 319

n.s., non specified.

positive concentration were lowest in mussels. Among all samples in-vestigated, the percentage of positive samples was highest for scallops(22%).

3.2. Consumption of shellfish

The mean usual long-term intake of scallops, molluscs, musselsand oysters for the whole Belgian population was 1.21 g/day for mus-sels, 0.14 g/day for oysters and 0.10 g/day for scallops, and 0.04 g/dayfor molluscs. The mean portion sizes for consumption days only were108 g/day for mussels, 103 g/day for oysters and 56 g/day for scallops(Table 2).

3.3. Acute exposure to domoic acid

The median acute intake of domoic acid (consumption days only,medium bound approach) was 0.53 μg/kg bw/day, while at the 99thpercentile the intake was 17.16 μg/kg bw/day. The acute referencedose of 30 μg/kg bw/day was exceeded between the 99th percentileand 99.9th percentile. This means that less than 1% of the populationwould be at risk (Table 3). Scallops were the biggest contributor toacute intake (43.4%), followed by mussels (36%), molluscs (14.9%) andoysters (5.7%). At the upper 5% of the distribution, the contributionswere 62.9% for scallops, 21.9% for molluscs, 11.7% for mussels and 3.5%for oysters (Tables 4a, 4b).

3.4. Chronic exposure to domoic acid

The usual long-term intakes of domoic acid were 0.02, 0.05, and0.16 μg/kg bw per day at the 90th, 95th and 99th percentile, respec-tively, using the lower bound concentrations. The usual intakes ofdomoic acid were 0.05, 0.08, and 0.17 μg/kg bw per day at the 90th,95th and 99th percentile, respectively, using the medium bound con-centrations. When using upper bound concentrations the usual intakes

Percentile Domoic acid intake(μg/kg bw/day)

Uncertainty of percentiles of domoic acidintake (μg/kg bw/day) (confidence limits)

2.50% 25% 75% 97.50%

50 0.53 0.41 0.48 0.57 0.6290 1.61 1.36 1.48 1.74 2.0395 2.54 1.86 2.21 3.05 3.4799 17.16 6.18 13.82 24.79 41.9199.9 177.33 77.03 133.13 223.81 545.6499.99 545.64 199.77 357.12 545.64 545.64Mean 1.61 1.09 1.37 1.88 2.34Maximum 545.64 235.25 395.5 545.64 545.64

bw, body weight.Number of simulations (consumers): 100,000 out of 3244.Positive intakes: 100,000 out of 100,000 consumptions.

Table 4aSummary characteristics of the total acute intake distribution of domoic acid using themedium bound approach.

Food Relativecontribution todomoic acidintake (%)

Mean(μg/kgbw/day)

P97.5(μg/kgbw/day)

Uncertainty of percentilesof domoic acid intake(μg/kg bw/day)(confidence limits)

2.50% 25% 75% 97.50%

Molluscs n.s. 14.9 0.24 0 0 0 20.5 37.3Mussels 36 0.58 2.78 20.9 30.6 42.8 53.6Oysters 5.7 0.09 1.02 1.3 3.6 8.7 14.7Scallops 43.4 0.7 1.19 22.1 36.6 51.7 65.4

Table 4bSummary characteristics of the upper acute intake distribution, using the mediumbound approach. Characteristics per food of the upper 5.0% of the distribution corre-sponding to a total intake higher than 2.5439 μg/kg bw/day.

Food Relativecontribution todomoic acidintake (%)

Mean(μg/kgbw/day)

P97.5(μg/kgbw/day)

Uncertainty of percentilesof domoic acid intake(μg/kg bw/day)(confidence limits)

2.50% 25% 75% 97.50%

Molluscs n.s. 21.9 4.55 0 0 0 28.4 50.7Mussels 11.7 2.44 8.38 4.3 8.3 15.5 22.9Oysters 3.5 0.74 0.16 0 0.3 7.8 16.8Scallops 62.9 13.08 119.89 32.9 53 78.9 90.5

118 M. Andjelkovic et al. / Environment International 49 (2012) 115–119

of domoic acid were 0.07, 0.10, and 0.18 μg/kg bw per day at the 90th,95th and 99th percentile. When taking only consumption days into ac-count, usual intakes of domoic acidwere 5.24, 7.24, and 12.76 μg/kg bwper day at the 90th, 95th and 99th percentile using the upper boundconcentrations (data not shown). The tolerable daily intake (TDI) of0.075 μg/kg bw per day was exceeded between the 94th and the 95thpercentile for the total population usingmedium bound concentrations.

Regarding the species contribution to the total usual domoic acidintake, scallops contributedmost to the total usual domoic acid intakewhen using lower (68.5%) andmedium (45.6%) bound concentrationsbut mussels contributed most to domoic acid intake when upperbound concentrations were used (47.6%). However, oysters contrib-uted the least, independent of the approach used (4.9%, 6.5%, and7.3%, respectively). When using the lower bound concentrations thesecond contributor species were samples marked as molluscs (19.4%)followed by mussels (7.2%) whereas mussels were the second contrib-utor species to the intake when using medium bound concentrations(34.7%), followed bymolluscs (13.1%). When using upper bound con-centrations mussels contributed most to domoic acid intake (47.6%),followed by scallops (34.9%), molluscs (10.2%) and oysters (7.3%)(Fig. 1).

0% 20% 40%

lower bound

medium bound

upper bound

Fig. 1. Statistics of largest contributions to the total intake distribution of domoic acid

4. Discussion

During the period 2004–2009 the Belgian Food Control Programscreened shellfish samples for domoic acid. The percentage of the positivesamples was highest for scallops and molluscs. A similar study in Croatiarevealed that of those species in the period from 2006 to 08 domoicacid was the most prevalent in blue mussel (Mytilus galloprovincialis),followed by the European flat oyster (Ostrea edulis), the Mediterraneanscallop (Pecten jacobaeus) and proteus scallop (Flexopecten proteus)(Ujević et al., 2010).

After the first reported and severe poisoning with amnesic shell-fish toxins in 1987 in Canada there were other episodes with less se-vere outcomes. Reports presented levels of domoic acid in razor clamsand scallops even up to 230 and 2900 ppm, respectively (Ramsdell, 2007and references therein). According to these occasional cases EFSA statedthat less than 1% of the population would be at risk for exceeding theacute reference dose. Our study confirms this for Belgian population. Inorder to protect high consumers against acute effects of marinebiotoxins, the CONTAM Panel of EFSA identified 400 g of shellfishmeat as an appropriate estimate of a large portion size consumed inEurope to be used in the risk assessments. According to the CONTAMpanel for a 60 kg adult, the chance of exceeding a dietary exposure of1.8 mg sum DA, corresponding to the ARfD of 30 μg sum DA/kg bw, isabout 1%, when consuming shellfish currently on the Europeanmarket.The chance of exceeding the deterministic dietary exposure estimate of8 mg sum DA, corresponding to consumption of a portion of 400 gcontaining DA at the level of the current EU limit value of 20 mg sumDA/kg, is less than 0.1%. EFSA refers to DA as a sum of domoic acidand epi-domoic acid which may also be measured in samples.

With regard to the long-term intake of domoic acid in Belgium thetolerable daily intake may be exceeded between the 94th and the95th percentile for total population using medium bound concentra-tions. This estimate corresponds to the high consumers of shellfish prod-ucts. However the clinical consequences of the long-term exposure todomoic acid in humans are not known. There are animalmodels evaluat-ing certain symptoms like DA-induced epilepsy (Ramsdell and Muha,2011) but these are also pointing out to age-related vulnerabilities.Some groups, like pregnant and lactating women, newborns, individualswith impaired renal clearance may be more susceptible to the toxins re-quiring special attention when developmental studies and environmen-tal regulatory policies are considered {Grant, 2010 24862/id}.

Another important public health implication is the DA heat stability.Although cooking does not destroy the toxin, normal home cookingprocesses, such as boiling and steaming, could reduce the amount ofDA in shellfishmeat due to partial leaching of the toxin into the cookingfluids (Vidal et al., 2009; McCarron and Hess, 2006). The effect ofcooking on the domoic acid concentration and the toxin concentrationin different species may be various. Toxin concentration may decreasein the cockle butmay increase in soft tissue ofManila clamafter cooking(Vidal et al., 2009). These changes in toxin quantity may have an

60% 80% 100%

molluscs

mussels

scallops

oysters

intake in Belgium expressed by lower bound; medium bound and upper bound.

119M. Andjelkovic et al. / Environment International 49 (2012) 115–119

important impactwhen assessing the acute and chronic DA intakes pro-vided that sample pre-treatment may be different e.g. cooking of mus-sels in order to open them easier before analysing. This is not alwaysaccounted for in the risk assessment studies and safeguarding the con-sumer health.

Data collected for regulatory control of food are often productsof targeted sampling in the expected problematic and designatingharvesting areas, while ideal sampling for exposure assessments shouldbe random (Verger and Fabiansson, 2008). This is not always possibleand may be costly. Only an evidence based approach to risk assessmentlike only conducting research after outbreaks may be disadvantageousin comparison to a knowledge based approachwhere we react accordingto processing already existent data and predict emerging problems. Onthe other hand, food frequency surveys give insight into the frequencywith which the listed food items are consumed during a specific time pe-riod (e.g. daily, weekly) but usually they do not give the number of eatingoccasions (Vandevijvere et al., 2009; Illner et al., 2010). A 24-h recall orone or more days of dietary record are needed to provide more detailedinformation such as the source of food, time of day and the placewhere foods are consumed. Dietary information can be collected atthree different levels of aggregation: national food supply data, data atthe household level and data at the individual level (Verger andFabiansson, 2008).

5. Conclusions

The present study showed that domoic acid occurs in shellfish inthe Belgian food chain. From routine testing it was found that lessthan 1% of the Belgian population would be at risk of acute DA intoxica-tion. These would mostly be the high consumers of scallops. Althoughthe effects of long term exposure to domoic acid are not yet clear, thereis also a possibility of delayed or latent DA neurotoxicity. The presentstudy showed that in Belgium around 5% of the population may be ex-posed to unsafe DA doses over longer periods. Given this and the severityof DA toxicity more tight control programs and dose–response studiesare required.

Acknowledgements

The authors are grateful for financial support from and proof-reading of themanuscript by the Belgium Federal Agency for the Safetyof the Food Chain (FAVV-AFSCA). Data presented in this manuscriptbelong to FAVV-AFSCA. The FAVV-AFSCA cannot be held liable for dam-ages resulting from the use of the data nor the conclusions that could bedrawn from their treatment.

Special thanks goes to Jacques Lhermitte and Jean-Yves Micheletwho diligently prepared all the samples for the analyses.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.envint.2012.08.007.

References

Bates SS, Leger C, Keafer BA, Anderson DM. Discrimination between domoic-acid-producingand nontoxic forms of the diatom Pseudonitzschia pungens using immunofluorescence.Mar Ecol Prog Ser 2012;100:185–95.

Blanco J, Acosta CP, Bermúdez De La Puente M, Salgado C. Depuration and anatomicaldistribution of the amnesic shellfish poisoning (ASP) toxin domoic acid in the kingscallop Pecten maximus. Aquat Toxicol 2002;60:111–21.

Costa LG, Giordano G, Faustman EM. Domoic acid as a developmental neurotoxin.Neurotoxicology 2010;31(5):409–23.

de Boer WJ, Van der Voet H. MCRA, release 6. A web-based program for Monte Carlo RiskAssessment. Bilthoven, Biometris, RIKILT and RIVM. Biometris and RIKILTWageningenUniversity and Research Centre RIVM Centre for Substances and Integrated Risk Assess-ment; 2007.

De Vriese S, De Backer G, De Henauw S, Huybrechts I, Kornitzer K, Leveque A, et al. TheBelgian food consumption survey: aims, design and methods. Arch Public Health2005;63:1-16.

EFSA. Contam. Scientific opinion of the panel on contaminants in the food chain on arequest from the European Commission on marine biotoxins in shellfish — domoicacid. EFSA J 2009;1181:1-61. [2-7-0009].

Goldstein T, Mazet JA, Zabka TS, Langlois G, Colegrove KM, Silver M, et al. Novel symp-tomatology and changing epidemiology of domoic acid toxicosis in California sealions (Zalophus californianus): an increasing risk to marine mammal health. ProcR Soc B 2008;275:267–76.

Hinder SL, Hays GC, Brooks CJ, Davies AP, Edwards M, Walne AW, et al. Toxic marinemicroalgae and shellfish poisoning in the British isles: history, review of epidemiology,and future implications. Environ Health 2011;10:54.

Illner AK, Nöthlings U, Wagner K, Ward H, Boeing H. The assessment of individual usualfood intake in large-scale prospective studies. Ann Nutr Metab. 2010;56(2):99-105.

James KJ, Gillman M, Amandi MF, López-Rivera A, Puente PF, Lehane M, et al. Amnesicshellfish poisoning toxins in bivalve molluscs in Ireland. Toxicon 2005;46:852–8.

Kumar KP, Kumar SP, Nair GA. Risk assessment of the amnesic shellfish poison, domoicacid, on animals and humans. J Environ Biol 2009;30:319–25.

Lefebvre KA, Robertson A. Domoic acid and human exposure risks: a review. Toxicon2010;15:218–30.

McCarron P, Hess P. Tissue distribution and effects of heat treatments on the content ofdomoic acid in blue mussels, Mytilus edulis. Toxicon. 2006;47(4):473–9.

Mafra Jr LL, Bricelj VM, Fennel K. Domoic acid uptake and elimination kinetics in oystersand mussels in relation to body size and anatomical distribution of toxin. AquatToxicol 2010;100:17–29.

Maucher JM, Ramsdell JS. Domoic acid transfer to milk: evaluation of a potential routeof neonatal exposure. Environ Health Perspect 2005;113:461–4.

Mauriz A, Blanco J. Distribution and linkage of domoic acid (amnesic shellfish poisoningtoxins) in subcellular fractions of the digestive gland of the scallop Pecten maximus.Toxicon 2010;55:606–11.

Muha N, Ramsdell JS. Domoic acid induced seizures progress to a chronic state of epilepsyin rats. Toxicon 2011;57:168–71.

EC. Commission Regulation (EC) No 2074/2005. Commission Regulation (EC) No2074/2005 of 5 December 2005 laying down implementing measures for certainproducts under Regulation (EC) No 853/2004 of the European Parliament and ofthe Council and for the organisation of official controls under Regulation (EC) No854/2004 of the European Parliament and of the Council and Regulation (EC) No882/2004 of the European Parliament and of the Council, derogating from Regulation(EC) No 852/2004 of the European Parliament and of the Council and amendingRegulations (EC) No 853/2004 and (EC) No 854/2004.

Ramsdell JS. The molecular and integrative basis to domoic acid toxicity. In: Botana L.,editor. Phycotoxins: Chemisty and Biochemistry. Ames, IA, USA: Blackwell Publish-ing; 2007. pp. 223-50.

Ramsdell JS, Muha N. Domoic acid induced seizures progress to a chronic state of epi-lepsy in rats. Toxicon. 2011;57(1):168–71.

QuilliamMA. Analysis of diarrhetic shellfish poisoning toxins in shellfish tissue by liquidchromatography with fluorometric and mass spectrometric detection. J AOAC Int.1995;78(2):555–70.

Schultz IR, Skillman A, Woodruff D. Domoic acid excretion in dungeness crabs, razorclams and mussels. Mar Environ Res 2008;66:21–3.

Slimani N, Valsta L. EFCOSUM Group. Perspectives of using the EPIC-SOFT programmein the context of pan-European nutritional monitoring surveys: methodologicaland practical implications. Eur J Clin Nutri 2002;56(2):S63-74. [Suppl].

Slob W. Probabilistic dietary exposure assessment taking into account variability inboth amount and frequency of consumption. Food Chem Toxicol 2006;44:933–51.

Trainer VL, Wells ML, CochlanWP, Trick CP, Bill BD, Baugh KA, et al. An ecological studyof a massive bloom of toxigenic Pseudo-nitzschia cuspidata off the WashingtonState coast. Limnol Oceanogr 2009;54:1461–74.

Ujević I, Ninčević-Gladan Ž, Roje R, Skejić S, Arapov J, Marasović I. Domoic acid— a newtoxin in the Croatian adriatic shellfish toxin profile. Molecules 2010;15:6835–49.

Vandevijvere S, De Vriese S, Huybrechts I, Moreau M, Temme E, De Henauw S, et al. Thegap between food-based dietary guidelines and usual food consumption in Belgium,2004. Public Health Nutr. 2009;12(3):423–31.

Verger Ph, Fabiansson S. Recent progress in exposure assessment and its interactionwith the risk analysis process. Trends Food Sci Technol 2008;19:S92–8. [Suppl.].

Vidal A, Correa J, Blanco J. Effect of some habitual cooking processes on the domoic acidconcentration in the cockle (Cerastoderma edule) and Manila clam (Ruditapesphilippinarum). Food Addit Contam Part A Chem Anal Control Expo Risk Assess.2009;26(7):1089–95.