Content uploaded by Arpád Ambrus
Author content
All content in this area was uploaded by Arpád Ambrus on Feb 19, 2019
Content may be subject to copyright.
This article was downloaded by: [A. Zentai]
On: 29 February 2012, At: 00:38
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: Chemistry,
Analysis, Control, Exposure & Risk Assessment
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tfac20
Exposure of consumers to morphine from poppy seeds
in Hungary
A. Zentai a , J. Sali a , M. Szeitzné-Szabó a , I.J. Szabó a & Á. Ambrus a
a Hungarian Food Safety Office, Budapest 1097, Gyáli út 2–6, Hungary
Available online: 10 Nov 2011
To cite this article: A. Zentai, J. Sali, M. Szeitzné-Szabó, I.J. Szabó & Á. Ambrus (2012): Exposure of consumers to morphine
from poppy seeds in Hungary, Food Additives & Contaminants: Part A: Chemistry, Analysis, Control, Exposure & Risk
Assessment, 29:3, 403-414
To link to this article: http://dx.doi.org/10.1080/19440049.2011.636762
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 and Contaminants
Vol. 29, No. 3, March 2012, 403–414
Exposure of consumers to morphine from poppy seeds in Hungary
A. Zentai*, J. Sali, M. Szeitzne
´-Szabo
´, I.J. Szabo
´and A
´. Ambrus
Hungarian Food Safety Office, Budapest 1097, Gya
´li u
´t 2–6, Hungary
(Received 25 May 2011; final version received 27 October 2011)
Poppy seed-containing foods are popular dishes in Hungary and some other Central European countries.
The alkaloids of poppy are used in the production of medicines. Poppy seeds used as food may also contain
considerable amounts of alkaloids, which raises the question of food safety. Morphine, codeine, thebaine and
noscapine concentrations of poppy seed samples from the period 2001–2010 and consumption data from two
Hungarian surveys, carried out in 2003 and 2009, were evaluated. Exposure calculations were made for morphine
intake by both point estimate and probabilistic methods, and the uncertainty of the calculated values was
estimated. The point estimate for the acute consumer exposure, calculated using the 97.5th percentiles of
morphine concentration and of poppy seed consumption and taking into account the reduction of morphine
content by processing, was 78.64 mg (kg bw)
1
day
1
for adults, and 116.90 mg (kg bw)
1
day
1
for children. Based
on probabilistic estimations, the 97.5th and 99th percentile exposures ranged between 18.3–25.4 and 25.6–47.4 mg
(kg bw)
1
day
1
for adults, and between 32.9 and 66.4 mg (kg bw)
1
day
1
for children, respectively. As a no
observed effect level (NOEL) had not been established, the significance of exposure could not be assessed.
Keywords: natural toxicants – alkaloids; exposure – probability modelling; exposure assessment
Introduction
Foods made from poppy seeds are popular in
Hungary. However, their natural alkaloid content
can be of concern. The poppy plant (Papaver
somniferum L.) contains several alkaloids, for instance
morphine, codeine, noscapine and thebaine. These are
important substances in the production of medicines.
Poppy plant varieties are grown for industrial and
dietary purposes as well. The published literature
indicates that poppy seeds contain only very low or
non-detectable levels of alkaloids (Lachenmeier et al.
2010), but the seeds may be contaminated during
harvest. The poppy seeds from plants cultivated for
industrial alkaloid production in Hungary are allowed
for human consumption only after proper cleaning.
Occasionally, the poppy seed marketed as a food
ingredient could contain high amounts of the alkaloids.
During the past five years the poppy seed was the
subject of increased research, particularly regarding the
reduction of its morphine contents. This was the result
of a serious intoxication case of an infant in Germany
in 2005 (Lachenmeier et al. 2010).
The consumption of large amounts of highly
contaminated poppy seeds may lead to light-
headedness and enteroparesis in sensitive individuals.
The symptoms described are in agreement with the
toxicological actions of morphine. In particular,
reports indicate that morphine intake from poppy
seeds may be on the scale of therapeutic morphine
doses (Battilani et al. 2009). In Hungary, official
authorities have recorded data of illnesses caused by
poppy seed consumption since 1985. The symptoms
ranged from paleness to nausea, vomiting, diarrhoea,
headache, stomach cramps, trembling, flushing, a
bitter sensation, stupor, dizziness, drowsiness, shiver-
ing, depression and itchy skin. Ground raw poppy seed
was consumed in all cases. However, the amount eaten
was not recorded (personal communication by the
National Institute for Food and Nutrition Science
2006).
After ingestion, morphine is relatively quickly
absorbed from the gastrointestinal tract, mainly from
the upper small intestine and, to a lesser degree, from
the stomach. The maximum effect of morphine is
achieved about 30 min after ingestion. The effect of a
single dose lasts about 4–6 h (Federal Institute for Risk
Assessment 2005). Morphine is metabolised mainly in
the liver and it undergoes conjugation with glucuronic
acid principally at the 3-hydroxyl group. Morphine is
excreted in urine mainly as morphine-3-glucuronide. In
addition to the 3,6-diglucuronide, other minor metab-
olites that have been identified include normorphine
and the 3-ethereal sulphate (Battilani et al. 2009).
Secondary conjugation also occurs at the 6-hydroxyl
group to form the 6-glucuronide, which is pharmaco-
logically active, and to a limited extent the
*Corresponding author. Email: zentai.andrea@mebih.gov.hu
ISSN 1944–0049 print/ISSN 1944–0057 online
ß2012 Taylor & Francis
http://dx.doi.org/10.1080/19440049.2011.636762
http://www.tandfonline.com
Downloaded by [A. Zentai] at 00:38 29 February 2012
3,6-diglucuronide. The low bioavailability (20–40%) in
the case of oral morphine administration can be
attributed to elimination through metabolism in the
intestinal mucosa and liver (Battilani et al. 2009).
There is no internationally accepted reference point
for the toxicity of poppy seed alkaloids. At a national
level, the German Federal Institute for Risk
Assessment (BfR) estimated 6.3 mg (kg bw)
1
day
1
as
a provisional daily upper intake level of morphine from
poppy seeds in 2005, based on the lowest oral
therapeutic dose of 1.9 mg morphine (2.5 mg morphine
sulphate pentahydrate) per person and a safety factor
of 5 (Federal Institute for Risk Assessment 2005). The
lowest starting oral dose administered to adults with
normal kidney function for pain killing is 2 30 and
210 mg day
1
for patients with restricted renal
diuretic capacity (Embey-Isztin 1996). Taking into
account the lowest 10 mg single dose, it is about five
times higher than the oral therapeutic dose used in the
BfR study. In the European Union, only Hungary has
a national legislation (17/1999 Eu
¨M rendelet 1999) that
specifies the maximum permitted levels at 30 mg kg
1
for morphine, 40 mg kg
1
for morphine and noscapine,
20 mg kg
1
for noscapine, 20 mg kg
1
for codeine, and
20 mg kg
1
for thebaine in poppy seed.
According to several studies, proper cleaning,
washing, grinding and baking could reduce signifi-
cantly the alkaloid contents of poppy seeds. Sproll
et al. (2006, 2007) studied the effect of processing on
the morphine level of poppy seed and found that the
optimal treatment for reduction consists of washing,
drying and grinding. All washing treatments led to a
significant reduction of the morphine content with a
minimal removal of 48% and a maximum removal of
the total morphine content. With cold water, only
about 60% of morphine was removed. At temperatures
above 60C, a significantly higher proportion, around
90%, could be removed. The length of poppy seed
rinsing also increased the level of reduction. Grinding
reduced the morphine and codeine contents on average
by 24% and about 20%, respectively (Sproll et al.
2007). In another study, Sproll et al. (2006) reported
morphine losses of 34% 5% due to grinding. They
also found that during baking up to 135C the
reduction of morphine was relatively low (around
30%), but at 220C a reduction of 80–90% was
observed.
The present study presents the results of the
evaluation of the acute exposure of Hungarian poppy
seed consumers to morphine, calculated with both
point estimate and probabilistic methods, based on the
alkaloid content of poppy seed marketed in Hungary
during 2001–2010, and on consumption data from the
surveys conducted in 2003 (Rodler et al. 2005) and
2009 (Szeitz-Szabo
´et al. 2011), also taking into
account the effect of processing. The uncertainty of
the calculated values was also estimated.
Materials and methods
The results of morphine, thebaine, noscapine and
codeine analyses in 737 poppy seed samples were
obtained from the market surveys carried out by the
Hungarian authorities. The National Institute for
Food Hygiene and Nutrition provided 566 data
samples from 2001 to 2006, and 171 data samples
were supplied by the Central Agricultural Office for
2007–2010.
During the first study period the alkaloid content
was measured with thin-layer chromatography accord-
ing to Hungarian standard MSZ 690:1993. In this
method the morphine was extracted with chloroform-
isopropyl alcohol, purified with 0.1 mol l
1
HCl solu-
tion, separated and detected on silica GF
254
TLC plate.
The recovery rate was 80%. The LOQ for morphine
was 2 mg kg
1
, and it varied between 2, 3, 4, 5 and
8mgkg
1
for noscapine, codeine and thebaine. The
results were confirmed with HPLC-UV detection.
Morphine was determined with HPLC-MS/MS
during 2007–2010. The poppy seed test portions were
extracted with methanol containing 1% HCl by
shaking for 2 h. The filtered extracts were directly
injected into the HPLC column (C18 5 mm) and
detected with MS/MS. The typical method perfor-
mance parameters reported were: LOQ ¼1mgkg
1
,
LOD ¼0.2 mg kg
1
, recoveries of 78–93%, 70–94%,
51–96% and 84–101%, and the reproducibility relative
standard deviation (RSD) at around 30 mg kg
1
was
6.3%, 5.9%, 7.6% and 15.2% for morphine, codeine,
thebaine and noscapine, respectively. The accuracy and
selectivity of the method were tested by spiking poppy
seed with known alkaloid contents and hemp which
does not contain alkaloids. The alkaloid standards of
known purity were purchased from Sigma-Aldrich Co.
LLC (J. Do
¨mso
¨di, personal communication).
Consumption data obtained with 3-day dietary
record methods from the years 2003 and 2009 were
used for the assessment. In 2003, 1360 adult consumers
recorded 79 poppy seed consumption days (1.94%),
while 4992 consumers (age range between 0 and 101
years) reported 327 poppy seed consumption days
(2.18%) in 2009. Regarding consumption data of
children and adolescents of 1–18 years of age, infor-
mation was only available from the consumption
survey conducted in 2009, where 1010 young respon-
ders reported 85 poppy seed consumption days in total
(2.81%). In the 2009 survey there were data available
about the recipes for foods, so it was possible to
evaluate the data for processing effects. The poppy
seed was consumed in ground form on 65% of days
and only in 35% of the cases was it consumed in baked
cakes.
In view of the infrequent consumption of poppy
seed and the rapid decrease of the potential adverse
effects of its alkaloid content, only the acute exposure
404 A. Zentai et al.
Downloaded by [A. Zentai] at 00:38 29 February 2012
of consumers is a concern. For the estimation of acute
exposure, both the alkaloid concentration data and the
poppy seed consumption data (eaters only) obtained
during the two periods were considered separately, and
after combining them for the whole time period
(2001–2010 for alkaloid concentrations, and
2003 þ2009 for the poppy seed consumptions).
For point estimate, the exposure was calculated
using the 97.5th percentile of daily poppy seed
consumptions and the 97.5th percentile of morphine
concentrations (European Food Safety Authority
(EFSA) 2007).
As consumption data for ground poppy seed and
baked poppy seed were available from the 2009 survey,
the effect of processing on morphine intake could also
be studied. The processing factor (P
f
) was calculated as
the ratio of the concentration of morphine in processed
(ground or ground and baked) poppy seed and raw
poppy seed (Food and Agriculture Organisation
(FAO) 2009). As poppy seed is not washed as part of
commercial or bakery operations in Hungary, the
effect of washing was not considered. The P
f
of 0.71 for
grinding of poppy seed was calculated from the 34%
and 24% average losses of morphine (Sproll et al. 2006,
2007). Strudels and cakes are baked in Hungary
typically at 180C. The P
f
of 0.31 for the baking of
cakes was estimated using an MS Excel exponential
trend line-fitting function from the reduction of
morphine content observed at 135 and 220C (Sproll
et al. 2006) assuming the reported quadratic relation-
ship. The combined processing factors including grind-
ing for morphine content in baked poppy seed was
0.71 0.31 ¼0.22. The morphine intakes were reduced
by the calculated processing factors according to the
form of consumed poppy.
For probabilistic modelling, parametric distribu-
tions were fitted to the morphine concentrations and
consumption data applying the @RISK programme
package. The uncertainty of the fitted parameters was
not estimated. The relative frequency distributions of
the original data and the fitted distributions were also
visually evaluated. Parametric distributions giving the
best fit for both the morphine concentrations and
consumption data were used to generate 200,000 and
500,000 random values. The generated lognormal or
gamma distributions were expected to cover the
high experimental values and beyond those
observed. Morphine exposure of children and adoles-
cents, as a subset of the total population, was
calculated similarly.
In summary, the exposure was calculated with each
of the following methods, for comparison:
Method (a): All consumption data were multiplied by
all morphine concentration data.
Method (b): The 200,000 or 500,000 pairs of random
values of morphine concentrations and poppy seed
consumptions, drawn from the selected parametric
distributions, were multiplied with each other. This is a
parametric alternative to method (a).
Method (c): The combined original data populations
(737 morphine concentrations and 406 poppy seed
consumptions) were bootstrapped 10,000 times and the
resulting morphine concentrations were multiplied by
the resulting consumption values (10,000 iterations).
The cumulative relative frequencies of exposure were
calculated from each iterations, using the NIST (2011)
definition for percentile calculations. This non-para-
metric bootstrapping procedure (Efron and Tibshirani
1994) was used to quantify uncertainty due to the
limited sample sizes.
Methods (d1) and (d2): Morphine intake was calcu-
lated using the abovementioned processing factors for
ground and baked poppy seeds applying procedures
(b) and (c).
The poppy seed consumption in raw ground form
and in baked cakes showed distinctly different pat-
terns, therefore parametric distributions were sepa-
rately fitted on the two datasets. The exposures of
consumers to morphine derived from consuming
ground poppy seed and cakes containing poppy seed
were calculated separately from the resampled distri-
butions of consumption and morphine data. As the
separate consumption data were only available from
the 2009 survey, the lognormal distribution fitted on
2007–2010 morphine concentration data was used for
the calculations. In order to obtain representative
exposure values for the combined poppy seed con-
sumption, the proportion of consumption days of
ground poppy seed (65%) and ground and baked
poppy seed (35%) were taken into account and 325,688
and 174,312 random samples were drawn from the
generated ground and baked poppy seed consumption
data population, respectively. The same number of
concentration data were drawn from the generated
morphine population. The exposures calculated from
the two sets of data were evaluated separately and in
combination.
The results obtained with these different methods
were compared and evaluated together.
Results and discussion
Alkaloid levels in poppy seeds in Hungary
Altogether 737 poppy seed samples were analysed.
Morphine was detected in 736 out of 737 samples.
Codeine and thebaine were detected in 61.3% and
63.0% of the cases, respectively. Noscapine was
detected in 6.2% of the samples only. In more than
half of the samples three or four alkaloids were
detected together. Table 1 presents the summary of
the detectable alkaloids and their selected percentiles,
in the samples.
Morphine content of the samples varied to a great
extent, with median of 11 mg kg
1
and highest values
Food Additives and Contaminants 405
Downloaded by [A. Zentai] at 00:38 29 February 2012
of 222, 238, 267 and 533 mg kg
1
. The 97.5th and 95th
percentiles and the median values of morphine con-
centrations were calculated with Harrell–Davis
(Peter Craig, personal communication), and the bino-
mial (Hamilton et al. 2004) methods, and with MS
Excel (Table 2). The binomial method enabled one to
estimate the probability that a given ranked value is the
selected percentile of the population. The Harrell–
Davis and the binomial methods resulted in very
similar estimates, while the 97.5th percentiles obtained
with Excel were lower. Consequently, the average
values of the median, 95th and 97.5th estimates
obtained with the Harrell–Davis and binomial meth-
ods are given in Table 3 for the alkaloids detected in
poppy seeds. The average estimated 97.5th percentile
of morphine concentrations and the similarly estimated
97.5th percentile of poppy seed consumptions were
used when calculating the point estimates of morphine
intake.
The relative frequency distributions of morphine
concentrations measured during the two survey periods
were similar (Figure 1). The ratios of the 97.5th
percentile and median values of the morphine 2007–
2010 and 2001–2006 datasets were 1.064 and 1.055,
respectively. The ratios indicate that the morphine
concentrations measured with different methods
during 2001–2006 and 2007–2010 were not
Table 1. Number of poppy seed samples analysed and the percentage of alkaloids detected in the samples above the LOQ.
Years 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Sample number 80 77 173 55 82 99 71 54 25 21
Morphine, % detected 100.0 100.0 100.0 100.0 100.0 99.0 100.0 100.0 100.0 100.0
Median 7.92 16.99 10.38 9.48 15.05 10.93 11.14 12.88 12.70 8.08
97.5th percentile 68.12 300.01 82.40 41.01 89.63 128.80 137.87 110.41 27.63 26.69
Codeine, % detected 30.0 64.9 48.0 56.4 75.6 69.7 80.3 75.9 96.0 52.4
Median of codeine concentrations 0.30 2.00 0.30 2.00 2.00 2.00 2.00 1.70 2.00 1.10
97.5th percentile of codeine concentrations 10.13 33.95 21.10 8.60 29.98 20.55 25.28 12.40 4.28 3.50
Noscapine, % detected 3.8 5.2 4.0 0.0 0.0 3.0 9.9 18.5 32.0 19.0
Median of noscapine concentrations 0.30 0.30 1.30 1.30 1.30 1.30 0.20 0.20 0.20 0.20
97.5th percentile of noscapine concentrations 2.05 5.20 4.00 1.30 1.30 4.98 6.28 6.22 11.58 5.10
Thebaine, % detected 27.5 61.0 50.9 61.8 67.1 78.8 90.1 70.4 92.0 71.4
Median of thebaine concentrations 0.30 2.00 2.00 2.00 2.00 2.00 3.40 1.70 2.50 1.40
97.5th percentile of thebaine concentrations 10.00 46.20 22.00 15.55 22.00 16.40 42.33 15.03 7.16 4.00
Notes: Calculated 97.5th percentiles are only indicative values and should be interpreted with caution, as the limited sample
number does not make their precise calculation possible. The percentiles were calculated with MS Excel.
Table 2. Calculated 97.5th percentiles and medians of morphine data (mg kg
1
).
Year; number
of samples
97.5th percentile of concentration Median concentration
Excel Harrell–Davis
Binomial method
Excel Harrell–Davis
Binomial method
LCL Calculated value UCL LCL Calculated value UCL
2001–2006; 566 80.00 85.74 67.00 89.21 160.00 11.00 10.95 10.00 10.95 11.00
2007–2010; 171 52.70 90.67 2.90 95.40 4202.00
a
11.40 11.53 10.80 11.57 12.70
2001–2010; 737 80.00 83.57 67.00 86.09 141.00 11.00 10.99 11.00 11.00 11.00
Notes: Harrell–Davis, values calculated with Harrell–Davis method; LCL, lower limit of 95% confidence interval; UCL: upper
limit of 95% confidence interval.
a
Number of data points was not sufficient for the estimation of confidence limits accurately.
Table 3. Distribution characteristics of the alkaloid levels
(mg kg
1
) in all (2001–2010) samples.
Median
a
Average
b
P0.95
a
P0.975
a
Maximum
Morphine 11.0 18.7 57.6 84.8 533.0
Codeine 2.0 3.6 17.2 25.7 60.0
Noscapine 1.2 1.2 1.6 4.7 40.0
Thebaine 2.0 4.0 15.8 23.9 120.0
Notes:
a
The median, 95th and 97.5th percentiles are the
average of the estimates obtained with Harrell–Davis and
binomial methods.
b
Average concentrations were calculated with the reported
LOQ/6 values for non-detected results (European
Commission Directorate – General Health and Consumer
Protection 2003).
406 A. Zentai et al.
Downloaded by [A. Zentai] at 00:38 29 February 2012
substantially different. The Mann–Whitney U-test
confirmed that the difference was not significant
(p¼0.35) (McDonald 2009).
The medians and 97.5th percentile concentrations
of alkaloids calculated for the 10 years of data
separately (Table 1) did not indicate a clear trend
when changing of morphine content. The calculated
97.5th percentiles were lower in the last years (2009 and
2010), but they may be attributed to the smaller
number of samples analysed, as the probability of
detecting at least one value above the 97.5th percentile
is only 40–45% in the case of 21–25 samples. Based on
these results, the combination of morphine data of the
10 years is justified.
Consumption of poppy seeds in Hungary
In the 2003 food consumption survey, poppy seed
consumption occurred on 1.94% of all consumption
days. Considering the body weight of consumers, the
highest, 97.5th percentile and average daily consump-
tion were 1.0, 0.80 and 0.34 g (kg bw)
1
day
1
,
respectively.
Poppy seed consumption was recorded on 327 out
of 14,976 days (2.18%) in the 2009 food consumption
survey. The average and 97.5th percentile consump-
tions were 0.44 and 1.20 g (kg bw)
1
day
1
, respec-
tively. The highest adult consumption was recorded by
a 20-year-old, 62-kg man (1.68 g (kg bw)
1
day
1
). The
average and 97.5th percentile consumptions of children
and adolescents were 0.67 and 1.77 g (kg bw)
1
day
1
,
respectively. The highest poppy seed consumption was
recorded by a 4-year-old, 22-kg girl (2.73 g (kg bw)
1
day
1
); while the youngest poppy seed consumer was 1
year old. It should be pointed out that none of the
consumption survey periods included Christmas, when
cakes are traditionally prepared with poppy seed.
The relative and cumulative frequency distributions
of 2003 and 2009 consumption data are shown
in Figure 2. The characteristics of the distributions
are given in Table 4. The difference in the consumption
figures might be attributed to the relatively small
number of consumption days in 2003.
There was detailed information on the form of
consumed poppy seed from the 2009 survey only. Out
of the 327 consumption days, pasta with ground, raw
poppy seed was consumed on 213 (65%) occasions,
and baked cakes were consumed on 114 (35%)
occasions. Consumption levels of raw, ground poppy
seed were generally higher than those of baked poppy
(Table 4).
Morphine intake from the consumption of poppy
seeds
Point estimate
Following the FAO/WHO recommended method
(World Health Organization (WHO) 1997) and the
current practice of the European Food Safety
Authority (EFSA) (2011), the 97.5th percentile of
poppy seed consumptions of eaters and the 97.5th
percentile of morphine concentration data were used
for calculating point estimates of acute intake.
In order to assess the potential consequence of
morphine concentrations and consumption values
obtained during the two periods, the intakes were
calculated from the two consumption datasets sepa-
rately and from the combined consumption data as
well as the separate and combined morphine concen-
tration datasets. The results are summarised in Table 5.
Figure 1. Relative frequency distributions of morphine concentrations detected in poppy seed.
Food Additives and Contaminants 407
Downloaded by [A. Zentai] at 00:38 29 February 2012
Figure 2. (a) Distributions of poppy seed consumption data in 2003 and 2009; and (b) distribution of poppy seed consumption in
ground raw and baked form in 2009.
Table 4. Poppy seed consumption (g (kg bw)
1
day
1
) based on the 2003 and 2009 surveys.
Survey data
2009
2003 2003 þ2009 Ratio
a
All Ground Baked
Number of days 327 213 114 79 406
Minimum 0.042 0.048 0.042 0.031 0.031
Average 0.440 0.516 0.300 0.340 0.421 1.294
Maximum 2.727 2.727 1.677 1.000 2.727
SD 0.304 0.294 0.271 0.183 0.287
CV 0.689 0.570 0.902 0.538 0.681
Median
b
0.395 0.439 0.217 0.337 0.382 1.172
P0.975
b
1.255 1.235 0.954 0.862 1.190 1.456
Notes:
a
Ratio of 2009/2003 consumption based on food survey data.
b
Median and P0.975 estimates were obtained with Harrell–Davis and binomial methods.
408 A. Zentai et al.
Downloaded by [A. Zentai] at 00:38 29 February 2012
Based on consumption of raw poppy seed, the
point estimate (using average percentiles calculated by
the binomial and Harrell–Davis calculation methods)
for the acute consumer exposure ranges between 73.1
and 116.7 mg (kg bw)
1
day
1
. The lowest exposure
(73.1 mg (kg bw)
1
day
1
) was obtained from the
combination of the 2003 consumption data and all
morphine concentrations, the highest exposure
(116.7 mg (kg bw)
1
day
1
) resulted from the com-
bination of 2009 consumption and 2007–2010 mor-
phine data.
The approximate 95% confidence limits for the
estimated intakes were calculated using the binomial
method from the ranked values bracketing the 68%
confidence intervals ((1 0.683)/2 ¼0.1587;
0.1587 0.1587 ¼0.025) of the 97.5th percentiles of
the corresponding datasets estimated based on the
binomial method (Figure 3). The standard deviation
(SD) of the selected percentile ‘p’ is calculated as:
SD ¼ffiffiffiffiffiffiffiffiffiffiffiffi
ðNpqÞ
p
where Nis the number of data points; and pþq¼1
(Diem and Seldrup 1982). The 68% confidence inter-
vals around the estimated 97.5th percentile for N¼737
correspond to the 17 4¼13th and 18 þ4¼22nd
ranked values (67 and 112 mg kg
1
), where the largest
value has first rank. As the best estimate (i.e. the
average of two calculation methods) of the 97.5th
percentile was close to the value estimated with the
binomial method, the confidence intervals calculated
with the values obtained from the binomial method
provide an acceptably close approximation. Due to the
low number of data points, the 95% confidence limits
for the 2003 consumption data and 2007–2010 mor-
phine concentrations encompassed practically the
whole datasets. Therefore, confidence limits could
not be calculated for those combinations.
Morphine short-term intakes of children and ado-
lescents (1–18 years) calculated from the 2007–2010
and 2001–2010 morphine concentrations and the 85
consumption days data of raw poppy seed were 164.7
and 150.2 mg (kg bw)
1
day
1
, respectively. The mor-
phine intake was higher than that of the whole
population due to the higher average and 97.5th
percentile of the poppy seed consumption of the age
group of 1–18 years. The small number of data did not
allow the calculation of the confidence intervals for the
estimated intakes to be made.
Probabilistic estimation of morphine intake
Probabilistic modelling was carried out with option
methods (a), (b), (c) and (d) described above. The
results are summarised in the following.
Table 5. Point estimate with 95% confidence intervals of acute exposure to morphine (mg(kg bw)
1
day
1
) from raw poppy seed
consumption.
Best estimates
a
(and 68% confidence intervals)
b
for morphine P0.975 concentrations in time periods (mgg
1
)
Best estimates
a
(and 68% confidence intervals)
b
for
P0.975 consumptions in time periods (g (kg bw)
1
day
1
)
2001–2006
87.48
(67.00–120.00)
2007–2010
93.04
2001–2010
84.83
(67.0–112.0)
Acute exposure
2003 0.86 75.42 80.21 73.14
2009 1.26 (1.18–1.33) 109.73 (78.79–159.96) 116.70 106.41 (78.79–149.30)
2003 þ2009 1.19 (1.11–1.32) 104.02 (74.44–158.88) 110.63 100.87 (74.44–148.29)
Notes:
a
Best estimate calculated as the average of estimated 97.5th percentile with Harrell–Davis and binomial method.
b
The 68% confidence interval calculated as P0.975 SD, as shown in Figure 3. The number of 2003 consumption figures and
2007–2010 morphine results were not sufficient for calculation of the confidence intervals.
0
1
2
3
4
5
6
7
8
9
0 5 10 15 20 25 30 35 40
Prob ability [%]
Ranked mor
p
hine concentrations:1= lar
g
est
p(x).x
Figure 3. Probability density function of the 97.5th percen-
tile of all morphine concentrations (2001–2010 data).
Explanation for the confidence limits: the 97.5th percentile
lies between the 17th and 18th ranked values from largest to
smallest. The 68% of the values (median SD) are between
the 17 4¼13 and 18 þ4¼22 ranked values.
Food Additives and Contaminants 409
Downloaded by [A. Zentai] at 00:38 29 February 2012
.The 200,000 values generated from the fitted
lognormal distributions covered the maximum
recorded consumption figures in each dataset;
however, the maximum experimental mor-
phine concentration could only be covered in
two out of three cases (columns B, C and D1
in Table 6). When 500,000 values were gener-
ated, the highest values obtained were sub-
stantially higher than the highest experimental
value (column D2). The results suggest that
the heavy tail of the empirical consumption
data can only be covered if a large number of
data points are generated from the fitted
distribution.
.The random morphine values generated from
the fitted gamma distribution, however,
underestimated the upper tail of the morphine
experimental data (the highest generated
and experimental values were 180.1 and
202, or 191.2 and 533 mgg
1
, respectively).
Consequently, the gamma distribution was
not suitable for estimating high quantiles of
exposure.
.When random samples were drawn from the
fitted lognormal distributions, the intake esti-
mates obtained from the different datasets
were in the same range below the 95th
percentile, but showed larger differences at
Table 6. Estimated cumulative distribution function (ECDF) of morphine intake from the consumption of poppy seed
calculated with probabilistic modelling.
ECDF A B C D1 D2 E FG HIJ
0.2 1.7 1.4 1.9 1.7 1.6 0.7 1.7 0.6 2.5 2.9 1.4
0.3 2.4 2.0 2.7 2.4 2.3 1.1 2.5 1.1 3.6 4.1 2.3
0.4 3.2 2.7 3.5 3.2 3.0 1.7 3.4 1.7 4.9 5.3 3.2
0.5 4.1 3.5 4.5 4.2 4.0 2.5 4.3 2.4 6.5 6.7 4.2
0.6 5.1 4.6 5.8 5.5 5.3 3.3 5.4 3.1 8.7 8.6 5.4
0.65 5.8 5.2 6.6 6.3 6.1 3.9 6.1 3.6 10.1 9.5 6.1
0.7 6.6 6.1 7.5 7.3 7.2 4.5 6.9 4.2 11.8 10.9 7.0
0.8 9.2 8.4 10.1 10.2 10.2 6.3 9.7 5.9 16.8 15.0 9.4
0.9 15.4 13.1 15.2 16.0 16.6 9.7 16.1 9.9 27.6 24.6 14.4
0.95 25.0 18.7 21.2 23.3 24.7 13.7 26.1 16.2 41.5 39.3 21.7
0.975 38.8 25.6 28.4 32.0 34.7 18.3 40.9 25.4 59.1 63.3 32.9
0.99 72.6 36.8 39.7 46.5 52.4 25.6 76.2 47.4 88.8 112.0 66.4
0.999 222.1 76.7 78.9 101.7 121.8 50.4 224.4 151.4 204.7 320.0 166.0
0.9999 489.1 147.4 131.9 195.9 235.5 84.9 499.7 336.9 398.1 702.7 321.7
*
1
1.74 4.37 4.67 10.61 4.67
*
2
1 2.73 2.73 2.73 2.73
*
3
498.4 250.6 569.4 719.8 719.8
*
4
533 202 533 533 533
Notes: Exposure results from processed poppy seed are underlined.
Explanation: 1: Poppy seed consumption (g (kg bw)
1
day
1
): maximum value of 200,000 or 500,000 generated data with the
parameters providing the best fit; 2: poppy seed consumption (g (kg bw)
1
day
1
): maximum experimental value; 3: morphine
concentration (mgg
1
): maximum values of 200,000 or 500,000 generated data with the parameters providing the best fit; and 4:
morphine concentration (mgg
1
): maximum experimental value.
Method (a), unprocessed poppy: Column A: multiplying all experimental 2003 þ2009 poppy consumption data (g (kg bw)
1
day
1
) with all (2001–2010) morphine data (mgg
1
) (406*737 data).
Method (b), unprocessed poppy: Column B: ln (2003 poppy seed consumption (g (kg bw)
1
day
1
)) and ln (morphine 2001–2006
(mgg
1
)), 200,000 data populations; Column C: ln (2009 poppy seed consumption (g (kgbw)
1
day
1
)) and ln (morphine 2007–
2010 (mgg
1
)), 200,000 data populations; Column D1: ln (2003 þ2009 poppy seed consumption (g (kg bw)
1
day
1
)) and ln
(morphine 2001–2010 (mgg
1
)), 200,000 data populations; Column D2: ln (2003 þ2009 poppy seed consumption (g (kg bw)
1
day
1
)) and ln (morphine 2001–2010 (mgg
1
)), 500,000 data populations.
Method (b), processed poppy: Column E: ln (2009 poppy seed consumption (g (kg bw)
1
day
1
)) and ln (morphine 2007–2010
(mgg
1
)), 500,000 data populations.
Method (c), unprocessed poppy: Column F: median of percentile exposure values obtained with bootstrapping of all
experimental 2009 poppy consumption data (g (kg bw)
1
day
1
) and 2001–2010 morphine data (mgg
1
).
Method (c), processed poppy: Column G: median of percentile exposure values obtained bootstrapping of experimental 2009
poppy consumption data (g (kg bw)
1
day
1
) with 2001–2010 morphine data (mgg
1
).
Children, method (b), unprocessed poppy: Column H: ln (2009 children poppy seed consumption (g (kg bw)
1
day
1
)) and ln
(morphine 2001–2010 (mgg
1
)), 500,000 data populations.
Children, method (a), unprocessed poppy: Column I: multiplying all experimental 2009 poppy consumption data of children with
all (2001–2010) morphine data (mgg
1
) (85*737 data).
Children, method (c), processed poppy: Column J: median of percentile exposure values obtained bootstrapping of experimental
2009 children poppy consumption data (g (kg bw)
1
day
1
) with 2007–2010 morphine data (mgg
1
).
410 A. Zentai et al.
Downloaded by [A. Zentai] at 00:38 29 February 2012
or above the 95th percentile of the cumulative
relative frequency distributions (Table 6, col-
umns B, C, D1 and D2). The difference
between the cumulative relative frequencies
corresponding to the intakes obtained with the
three combinations of data represents the
uncertainty of the exposure estimation based
on the calculations with the limited subsets of
data.
.The results obtained by multiplying all the
experimental data and bootstrapping
(methods (a) and (c)) are much higher than
those obtained by method (b) and therefore
represent worst-case situations (see columns A
and D2, and columns H and I).
.The comparison of columns C and E illus-
trates the effect of processing based on the
2009 consumption and 2007–2010 morphine
concentrations. The ratio of exposure from
raw poppy seed and from processed poppy
seed in the percentiles above 95th is between
0.639 and 0.646, which indicates that the
grinding (P
f
¼0.71) was the main source of
reduction of exposure.
.The comparison of columns F and G illus-
trates the effect of processing on exposure
based on 10,000 bootstrapped data (the expo-
sure percentiles were calculated from each
iteration and the medians are shown in Table
6). The ratios of corresponding percentiles are
between 0.621 and 0.675 at the higher
percentiles.
.The exposure of children (mg (kg bw)
1
day
1
)
to poppy seed shown in columns H, I and J is
about one to two times higher than of adults
(columns D2, A and G) at the 97.5th to
99.99th percentiles.
The cumulative distribution of the exposure calcu-
lated by multiplying the morphine concentration and
consumption data obtained from their fitted lognormal
distributions and by multiplying all of the experimental
data (method (a)) are close to each other up to about
the 90th percentile of the exposure (columns A, D1 and
D2). At the 99.9th and 99.99th percentiles, however,
the differences are more than two times, even in the
cases where the highest modelled values for poppy seed
consumption and morphine concentrations are higher
than the corresponding experimental values. The
possible reason for the large differences at high
percentiles was further studied. The highest morphine
concentrations observed in poppy seed samples were
187, 200, 202, 222, 238, 267 and 533 mg kg
1
, and the
highest poppy seed consumptions were 1.30, 1.32, 1.33,
1.54, 1.58, 1.68 and 2.73 g (kg bw)
1
day
1
. It was
found that the four highest morphine concentrations
in combination with the highest 20 poppy seed
consumption values resulted in exposures above the
99.99th percentile cumulative exposure (489.1 mg
(kg bw)
1
day
1
). If the highest morphine and poppy
seed consumption values (533 mg kg
1
and 2.73 g
(kg bw)
1
day
1
) were omitted from the calculation,
then the 99.99th percentiles would be 267 mg (kg bw)
1
day
1
. These results explain why lower exposure values
were obtained when they were calculated from the
results of random sampling of fitted populations
(columns D1 and D2 in Table 6), because in the
latter cases the probability of selecting the highest
values from both populations of 200,000 or 500,000
with random sampling is very low.
The range of estimated cumulative relative fre-
quencies was calculated by bootstrapping, drawing
10,000 random samples with replacement from both
the combined 737 morphine concentrations and the
406 poppy seed consumption data, and multiplying all
concentrations with all consumption data in each case
to calculate the morphine intake (mg (kg bw)
1
day
1
).
The procedure was repeated 10,000 times (bootstrap
method (c)). The median and 95% confidence limits of
the calculated cumulative relative frequencies are
shown in Figure 4. The horizontal lines within the
confidence limits show the intake range belonging to a
given cumulative percentile. The vertical line between
the confidence intervals indicates the range of the
percentage of the consumption days of morphine
intake around the selected one. The results showed
that the 95% confidence interval of the estimated
cumulative frequencies, indicating the uncertainty of
the estimation, was the narrowest at the 50th percentile
and increased at higher and lower percentiles. For
instance, the relative 95% interval of the estimated
50th, 95th, 97.5th, 99th, 99.9th and 99.99th cumulative
percentiles were in the order of 13.2%, 27.7%, 40.1%,
50.3%, 69.7% and 80.5%.
Figure 4. Median and 95% confidence intervals of the
exposure to morphine through consumption of poppy seed.
Food Additives and Contaminants 411
Downloaded by [A. Zentai] at 00:38 29 February 2012
Figure 5 shows examples for the frequency distri-
butions of estimated percentiles obtained from 10,000
bootstrap samples. At or below the 99th percentile the
calculated values approximate a normal distribution.
However, at higher percentiles the distribution is
becoming scattered and their ranges become wider, as
expected.
Effect of processing on morphine intake
The calculations, assuming that the morphine concen-
trations measured in raw poppy seed are present in
consumed food, provide an overestimate of the intake.
In order to obtain a more realistic estimate, the effects
of washing, grinding and baking were taken into
consideration. In Hungary, poppy seeds produced on a
commercial scale are not washed. Taking into account
that 65% of poppy seeds are consumed unbaked, and
35% are consumed baked, the acute intake of mor-
phine was recalculated by applying the calculated
processing factors (0.71 and 0.22). The estimated
exposure decreased by 32.6% for adults and by
29.0% for children, respectively (from 116.70 to
78.64 mg (kg bw)
1
day
1
and from 164.70 to
116.90 mg (kg bw)
1
day
1
respectively).
Regarding the probabilistic exposure estimate of
processed poppy seed consumption, it was found that
the intake was decreased by about 33–38% at the
upper percentiles. These findings are in accordance
with the higher intakes of ground poppy seed (the
estimated processing factor is 0.71) than that
consumed in cakes, and consequently ground poppy
seed consumption is the cause of the high morphine
intake.
Counting for a consumer of 60 kg (body weight),
the starting oral therapeutic dose reported by
Embey-Isztin (1996) would be about 167 mg
(kg bw)
1
, while the lowest oral therapeutic dose
reported by the BfR would be about 31.7 mg
(kg bw)
1
. The results from probabilistic calculations
suggest that the 31.7 mg (kg bw)
1
dose could be
reached only above the 97.5th percentile of the adult
poppy seed eaters (Table 6, columns B, C, D1 and D2).
Looking at children’s exposure (Table 6, column H),
above the 90th–95th percentiles could be in the range
of the estimated lowest therapeutic dose referred by the
BfR. The probability of reaching the starting oral
therapeutic dose of 167 mg (kg bw)
1
is very low and
practically negligible.
The provisional daily upper intake level (6.3 mg
(kg bw)
1
day
1
) established by the BfR has an incor-
porated uncertainty factor of 5. This is not likely to be
exceeded by the median adult exposure; however,
children have higher intake than adults. Around 65%
of the cumulative distribution of morphine intake of
adults and 50% of children corresponds to the 6.3 mg
(kg bw)
1
day
1
provisional daily upper intake calcu-
lated with the morphine content of unprocessed
poppy seed.
Taking into account that poppy seed is consumed
in processed form, it was concluded that exposures are
about 35% lower than the scenarios calculated without
counting for processing losses.
Figure 5. Relative frequency of estimated quantile values calculated from 10,000 bootstrap. Solid line: normal distribution fitted
on results; dashed line: lognormal distribution fitted on results.
412 A. Zentai et al.
Downloaded by [A. Zentai] at 00:38 29 February 2012
By applying the point estimation methodology, the
calculated acute intake exceeds the BfR reference
value, but it does not reach the starting oral therapeu-
tic dose. It should be noted that this is a very
conservative intake calculation; the probability that a
‘big eater’ consumes from the highly contaminated
food is low.
Conclusion
The most prominent alkaloid of poppy seed is
morphine, which was present in detectable amounts
practically in all samples analysed in Hungary during
2001–2010. As the poppy seed was consumed in only a
small fraction of the dietary survey days, the chronic
intake is not of a concern. Consequently only the
short-term intakes of poppy seed alkaloids were
estimated in our study, which was based on the
measured alkaloid content of raw poppy seed, also
taking into account the estimated effect of grinding
and baking. The intake calculation performed with the
fitted lognormal distributions gave the best fit on the
original data. The 97.5th percentile exposure of adults
ranged from 25.6 to 34.7 mg (kg bw)
1
day
1
depending
on the time period. The calculated 97.5th percentile of
children’s exposure was 59.1 mg (kg bw)
1
day
1
.
Taking into account the reduction of morphine content
by processing, the 97.5th and 99th percentile exposures
based on probabilistic estimations ranged between 18.3
and 25.4 and between 25.6 and 47.4 mg (kg bw)
1
day
1
for adults, and between 32.9 and 66.4 mg (kg bw)
1
day
1
for children, respectively.
The poppy seed was used on 65% of consumption
days as flavouring ingredient of pasta in ground form.
The average and 97.5th percentile consumptions of
poppy seed in baked form were 60% and 77% of that
eaten in ground form, respectively. The high morphine
intake primarily derives from the consumption of
ground raw poppy seed. According to current market-
ing practice, the poppy seed is cleaned only by venti-
lation and screening, but is not washed. Consequently,
the morphine content of the commercialised raw poppy
seed is reduced before consumption by about 35% due
to the effect of grinding. The consequences of the
pattern and form of consumption are reflected in the
results of estimated short-term exposure which indi-
cated that the morphine intake from processed poppy
seed was about 35% lower than the values calculated
from the morphine content of raw poppy seed.
The five-fold difference in reported lowest oral
therapeutic doses of 167 and 31.7 mg (kg bw)
1
of
morphine indicates the large uncertainty in the esti-
mation of the reference dose value for assessing the
potential adverse affects of exposure of consumers to
morphine.
Reliable risk assessment was not possible because
internationally accepted acute reference doses for
morphine and the other alkaloids detected in poppy
seeds have not yet been established. Even less or no
information is available on the potential adverse effects
of the codeine, thebaine and noscapine content of
poppy seeds individually or in combination, and the
bioavailability of these alkaloids.
Further scientific studies, meeting the relevant
standards, are required in order to establish the acute
reference dose in order to assess the actual risk. The
effect of technological steps, the degradation of mor-
phine in the digestive system during absorption, and
the effect of the poppy alkaloid ‘cocktail’ compared
with purified morphine should also be taken into
account. Above that, the necessary risk management
options could be established to reduce the alkaloid
intake. Our findings contribute to the scientific risk
assessment by providing exact, statistically well-
founded exposure data of poppy seed alkaloids, with
a specific focus on morphine.
Acknowledgements
The authors are grateful to the former National Institute for
Food Hygiene and Nutrition (now named the National
Institute for Food and Nutrition Science) and the Central
Agricultural Office for providing the results of alkaloid
analysis in poppy seed; and to Peter Craig, Durham
University, for providing the Excel template for the calcu-
lation of percentiles with the Harrell–Davis method.
References
17/1999 Eu
¨M rendelet. 1999. 17/1999 (VI. 16.) Eu
¨M rendelet
az e
´lelmiszerek vegyi szennyezettse
´ge
´nek megengedheto
00
me
´rte
´ke
´ro
00 l; [cited 2011 Oct 21]. Available
from: http://www.complex.hu/jr/gen/hjegy_doc.cgi?
docid=99900017.EUM/
Battilani P, Costa LG, Dossena A, Gullino ML, Marchelli R,
Galaverna G, Pietri A, Dall’Asta C, Giorni P, Spadaro D,
et al. 2009. Scientific information on mycotoxins and
natural plant toxicants. Scientific/Technical Report sub-
mitted to EFSA. Parma: European Food Safety Authority
(EFSA); [cited 2011 Oct 21]. Available from: http://
www.efsa.europa.eu/en/supporting/doc/24e.pdf/
Diem K, Seldrup J. 1982. Introduction to statistics.
Statistical tables mathematical formulae. In: Lenter C,
editor. Geigy scientific tables. Vol. 2. Basel (Switzerland):
Ciba–Geigy. p. 219.
Efron B, Tibshirani R. 1994. An introduction to the
bootstrap. Boca Raton (Florida): Chapman & Hall/CRC.
Embey-Isztin D. 1996. Analgesic treatment of patients with
tumor. Ha
´ziorvos Tova
´bbke
´pzo
00 Szemle 1:49–50; [cited
2011 Oct 21]. Available from: http://sunrise.sote.hu/htsz/
embey.htm/ (in Hungarian).
European Commission Directorate – General Health and
Consumer Protection. 2003. Collection of occurrence data
of Fusarium toxins in food and assessment of dietary
intake by the population of EU Member States, Report of
experts participating in Task 3.2.10. April 2003. Brussels
Food Additives and Contaminants 413
Downloaded by [A. Zentai] at 00:38 29 February 2012
(Belgium): European Commission; [cited 2011 Oct 21].
Available from: http://ec.europa.eu/food/fs/scoop/
task3210.pdf/
European Food Safety Authority (EFSA). 2007. Reasoned
opinion on the potential chronic and acute risk to
consumers health arising from proposed temporary EU
MRLs. EFSA; [cited 2011 Oct 21]. Available from: http://
www.efsa.europa.eu/en/efsajournal/pub/32r.htm/
European Food Safety Authority (EFSA). 2011. EFSA
model for chronic and acute risk assessment – rev. 2_0.
EFSA; [cited 2011 Oct 21]. Available from: www.efsa.
europa.eu/en/mrls/docs/calculationacutechronic_2.xls/
Federal Institute for Risk Assessment. 2005. BfR recom-
mends provisional daily upper intake level and a guidance
value for morphine in poppy seeds. BfR Health
Assessment No. 012/2006.
Food and Agriculture Organisation (FAO). 2009. FAO
Manual on the submission and evaluation of pesticide
residues data for the estimation of maximum residue
levels in food and feed. Rome (Italy): FAO; [cited 2011
Oct 21]. Available from: http://www.fao.org/fileadmin/
templates/agphome/documents/Pests_Pesticides/JMPR/
FAO_manual2nded_Oct07.pdf/
Hamilton DJ, Ambrus A, Dieterle RM, Felsot A, Harris C,
Petersen B, Racke K, Wong S-S, Gonzalez R, Tanaka K.
2004. Pesticide residues in food-acute dietary intake. Pest
Manag Sci. 60:311–339.
Lachenmeier DW, Sproll C, Musshoff F. 2010. Poppy seed
foods and opiate drug testing – where are we to day? Ther
Drug Monit. 32:11–18.
McDonald JH. 2009. Handbook of biological statistics;
[cited 2011 Oct 21]. Available from: http://udel.edu/
mcdonald/statkruskalwallis.html/
MSZ 690:1993. Determination of morphine content
in poppy seed for food purpose; 2003 2nd ed. Budapest
(in Hungarian).
NIST. 2011. NIST/SEMATECH e-Handbook of Statistical
Methods; [cited 2011 Oct 21]. Available from: http://#160
www.itl.nist.gov/div898/handbook/prc/section2/prc252.htm/
Rodler I, Bı
´ro
´L, Greiner E, Zajka
´s G, Szo
´ra
´d I, Varga A,
Domonkos A, A
´goston H, Bala
´zs A
´, Mozsa
´ry E, et al.
2005. Ta
´pla
´lkoza
´si vizsga
´lat Magyarorsza
´gon, 2003–2004.
Orvosi Hetilap. 146(34):1781–1789.
Sproll C, Perz RC, Buschmann R, Lachenmeier DW. 2007.
Guidelines for reduction of morphine in poppy seed
intended for food purposes. Eur Food Res Technol.
226:307–310.
Sproll C, Perz RC, Lachenmeier DW. 2006. Optimized
LC/MS/MS analysis of morphine and codeine in poppy
seed and evaluation of their fate during food processing
as a basis for risk analysis. J Agric Food Chem. 54:
5292–5298.
Szeitz-Szabo
´M, Biro
´L, Biro
´Gy, Sali J. 2011.
Dietary survey in Hungary, 2009. Part I: Macronutrients,
alcohol, caffeine, fibre. Acta Alimentaria. 40:142–152.
World Health Organization (WHO). 1997. International
estimated short-term intake (IESTI) and international
estimated daily intake (IEDI). Geneva (Switzerland):
WHO; [cited 2011 Oct 21]. Available from: http://
www.who.int/foodsafety/chem/acute_data/en/index1.html/
414 A. Zentai et al.
Downloaded by [A. Zentai] at 00:38 29 February 2012
