Вы находитесь на странице: 1из 9


ISSN: 0027-7878 (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/zfnr17

Food additatives. Use, intake and safety

Nils-Gunnar Ilbäck & Leif Busk

To cite this article: Nils-Gunnar Ilbäck & Leif Busk (2000) Food additatives. Use, intake and safety,
Näringsforskning, 44:1, 141-149, DOI: 10.3402/fnr.v44i0.1778

To link to this article: https://doi.org/10.3402/fnr.v44i0.1778

© 2000 The Author(s). Published by Taylor &


Published online: 13 Dec 2016.

Submit your article to this journal

Article views: 3236

Full Terms & Conditions of access and use can be found at

Scandinavian Journal of NutritionlNaringsforskning Vol44:141-149,2000

Food additives. Use, intake and safety

By Nils-Gunnar Ilback and Leif Busk

Today we have about 300 approved food additives on the European market and all of them have
been evaluated for safety by the European Commission's expert advisory body - the Scientific
Committee for Food (SCF). There are fundamental requirements for approval of a food additive.
The additive has to be of value for the consumer or necessary for the handling and distribution
of the food. In addition, it has to be toxicologically acceptable according to European Community
regulatory requirements. For most additives, a numerical (mg/kg bw/day) ADI-value (Accep-
table Daily Intake) has been determined that shows how much it is possible to consume each day
during a life time without any appreciable risk of adverse effects. If no adverse effects have been
demonstrated in the safety evaluation no numerical value for the AD1 is specified and the additive
may be used according to the principles of Good Manufacturing Practice (quantum satis), i.e. the
minimum amount necessary to produce the desired effects. When the intake of certain food
additives can be assumed to be high in relation to the ADI, this should always trigger additional
consideration and action. For such food additives it is important that continuous monitoring of
use and intake is performed, that potentially sensitive groups and/or individuals are identified and
that intake calculations are performed to ensure that the AD1 is not exceeded.
Key words: Consumption, food additives, intake, regulatory requirements, safety assessment

Introduction The use of food additives originated in cussed in the section "Safety evaluation of
The major goal for the National Food ancient times. Far back in time people a food additive". Furthermore, food
Administration (NFA) is to ensure the understood that some foods contained additives cannot be used in our food unless
safety of food. In order to achieve this, it is natural chemical compounds that could be they are of benefit for the consumer or
important to perform risk assessments of used in food storage, although the struc- needed for the handling of the food. Today
microorganisms and chemicals (contami- ture and mechanisms of action of the sub- we have about 300 approved food
nants and food additives) in food. The stances were unknown. For example, the additives and about 2500 flavouring
major reason for using food additives is to naturally occurring chemical substance agents on the market. The use of food
preserve and in different ways improve the benzoic acid (E210) is present in several additives is strictly regulated, but at pre-
quality of food. Food additives must be of berries and fruits, such as lingon berries, sent this is not the case for the flavouring
value for the consumer or necessary for and acts as a preservative and protects agents, which are presently being
the handling and distribution of the food against yeast, bacteria and fungi. Thus, evaluated within the European Union (see
and they have to be toxicologically accep- lingon berries have historically been used section "Flavouring agents").
table. In order to keep the intake of food to preserve fruit that could not be stored in
additives within safe levels it is important other ways. Food additives
to regulate the use of the additives in food, In previous times it was not possible to There are two major categories of food
as well as to estimate consumption of the get pure chemicals, but today we use either additives, enrichment substances and the
foods that contain the specific additive. synthetically produced food additives or technological additives. Enrichment sub-
additives extracted and purified from stances are used to improve the nutritional
natural sources with well-described mecha- value of food and have been used as a tool
Nils-Gunnar Ilback PhD. Toxicologist at National nisms of action. Since 1920, the use of in reducing deficiency diseases. For
Food Administration, Uppsala, and kssoc. Prof. at officially approved food additives has example, iodine is added to salt to avoid
Section of Infectious Diseases, Department of become common (1). However, in recent goitre and AD-vitamins are added to
Medical Sciences, Uppsala University Hospital,
Uppsala, Sweden. Leif Busk PhD. Head of R&D, decades the regulation of food additives margarine to avoid health effects due to
Research and Development Department, National has become a matter of controversy. Al- deficiency of these nutrients. However,
Food Administration, Uppsala, Sweden. though informed individuals generally deficiency diseases are not a major pro-
Correspondence: Nils-Gunnar Ilback, Toxico-
logy Division, National Food Administration, P.O. acknowledge that the quality and safety of blem in Europe today and the need for
Box 622, SE-75 1 26 Uppsala, Sweden. food have improved during the past 100 enrichment has thus become less. The ma-
E-mail: ngil@ slv.se years, some individuals disagree and a few jor area today is technological food addi-
disagree vigorously. Explanations for this tives and the present review article will
controversy, which is likely to continue, focus on the use and safety of these agents.
are easy to identify and will be discussed Technological food additives are mainly
in the following text. used to increase the shelf life of food
It is not permitted to use additives in (preservatives and antioxidants), or to
food unless they are pure and have been give it a better taste (sweeteners and fla-
evaluated for safety. The requirements for vouring agents), or to change its con-
the safety evaluation of food additives are sistency (emulsifiers and thickeners).
very similar to those for pharmaceutical Today we have about 300 technological
compounds and this will be further dis- food additives that are approved for use
Food additives. Use, intake and safety

and they can be divided into sub groups such as glucose, sucrose and fructose, and Table 1: Sweeteners used in Sweden, an example
(2). Most of the additives are present in another group of sugar alcohols, such as of a group of food additives that are of interest for
many consumers, not only diabetics. Sugars are
nature as natural constituents, of for sorbitol and xylitol (Table 1). The second not regarded as food additives.
example, fruits and vegetables. However, group contains the synthetic sweeteners
some of them are now produced syn- without energy, such as aspartame, ace- E n e r ~containing sweeteners
thetically. Other additives that are pro- sulphame K, saccharin and cyclamate. Sugars
duced synthetically are not present in This group of artificial sweeteners has Saccharose
nature. Semi-synthetic additives are caused a lot of debate regarding safety, Fructose
natural compounds that have been which will be discussed in a following Glucose
chemically modified. Nevertheless, all section "Alarm in the media". Lactose
these food additives, except for the Maltose
flavouring agents, are pure compounds Additional additives, such as pH regula- Sugar alcohols
that have been evaluated for safety and tors: In this group there are different acids, Sorbitol
their use and intake are regulated, as will bases and salts, for example glutamic acid Mannitol
be discussed in the present article. How- and its salts (E620-625) that are used as Xylitol
ever, there are presently no regulatory taste intensifiers. These are also additives Isomalt
rules for flavouring agents, but such rules that over the years have caused headlines Lactitole
in the media and public debate (section Maltitol
will be introduced in the near future
following an ongoing safety evaluation in "Alarm in the media"). Energy free sweeteners
the European Union (see section "Fla- Artificial sweeteners
Flavouring agents: This group of food Acesulfame K E 950
vouring agents"). additives is used to give the foods a special
The main groups of food additives and Aspartame E 951
aroma or taste. Flavouring agents can be Cyclamate E 952
their areas of use will be briefly described produced artificially or extracted from Saccharin E 954
in the following section: natural sources. Some of these artificially Natural sweeteners
Colouring agents and dyes: It is not per- produced compounds are present in nature, Taumatine E 957
mitted to use colouring agents and dyes in whereas others are not. They can be used Neohesperidin DC E 959
our staple foods. The reason is partly the either individually or in mixtures of flavour-
general restrictive attitude to additives and ing substances. This is a large group,
partly the fact that staple foods are con- approximately 2500 substances on the bitter almond derived flavouring agent,
sumed in high amounts. market, and so far their use has not been
regulated as strictly as other food additives the deadly dose is 6- 10 bitter almonds for
Preservatives: These additives are used to (see section below). a child and 40-60 almonds for an adult.
preserve and protect the food from inva- Thus, it is very important to evaluate and
sion and degradation induced by micro- regulate the use of all these naturally
organisms such as bacteria, fungi and Flavouring agents a very - occurring and sometimes dangerous
moulds. Preservation also gives protec- special group of food additives flavouring substances in the same way as
tion from natural toxins that may be pro- The flavouring agents are a very special for other food additives.
duced by microorganisms. In earlier days, and complex group of additives: since Within the EU there is an ongoing pro-
people used air-drying, smoke, salt and they occur naturally, and are not regulated, gramme to evaluate 3000 flavouring
fermentation as natural preservatives. they can not be regarded as "normal" food agents within 5 years (by the end of 2005).
Some of these methods are used even additives. Natural sources of flavourings The Council of Europe is presently eva-
today, but now we also have other methods, are materials of vegetable or animal ori- luating 600 natural sources of flavourings:
such as heat treatment, deep-freezing and gin, some of which are consumed as food 101 of these evaluations have been fina-
chemical preservation. (3). For most of these flavouring sub- lised to date (3). The safety evaluation
Antioxidants: These are used to protect fat stances and mixtures we need more infor- procedure used by EU, provides a scienti-
from becomingrancid and fruit containing mation about their toxicity, the amount fically based practical method of integrat-
food from discolouration. Vitamins (such that is used as food additives and in which ing data on intake, structure-activity
as A, D, E and B,) that are sensitive to foodstuffs they are actually used. relationships, metabolism and toxicity to
degradation can also be protected by the Some of the naturally occurring flavour- evaluate flavouring substances in a timely
action of antioxidants. ing agents may be toxic to humans in high manner. The procedure has been used
doses and are of more concern than the recently by the Joint FAOIWHO Expert
Emulsifiers-, stabilizing- and thickening other food additives that are well re- Committee on Food Additives (JECFA),
agents, including anti-caking agents: This gulated.For example, hydrogen cyanide is as an advisory board, to evaluate a total of
is an extensive group with several a poison well-known to readers of crime 263 flavouring substances (5). The pro-
different compounds intended to be used novels, it can also be a hazardous pollutant blem with a safety evaluation of mixtures
in order to affect the consistency of foods. in occupational exposure in industry, it of flavouring agents could be illustrated
Some compounds in this group of addi- can be present as a food additive in salt as by the coffee bean flavour that is used in
tives have caused concern among con- sodium ferrocyanide (E 535) and it can instant coffee, chocolate, ice cream, sweets,
sumers, for example the anti-caking agent also be a flavouring agent that is present soft drinks, etc. This flavour contains
sodiumferrocyanid (E 535) in salt (see naturally in bitter almonds and in stone more than 900 chemically defined sub-
section "Flavouring agents"). fruits (4). When used as a food additive, stances but only a limited and specifically
Sweeteners: There are two main groups of sodium ferrocyanide (E 535) does not de- selected number of them are used for the
sweeteners, the natural and the synthetic compose or release cyanide. As other "coffee flavouring mixtures" in foods.
(artificial) sweeteners. The natural sweete- regulated food additives it has been evalu- However, the composition of the flavour-
ners contain energy and can be further ated for safety and is of no concern. On the ing mixture is in most cases unknown,
subdivided in one group of general sugars other hand, when cyanide is ingested as a except to the industry. This illustrates how
Ilback and Busk

Figure 1. Toxicological data needed for safety Safety evaluation should be submittedby the petitioner. This
assessment and ADI. includes administrative data, technical
of a food additive
Increasing global trading has increased data (specification, manufacturing process,
Metabolism and Pharmacokinetics
the number of commercially available purity, methods of analysis, etc.) and
food products on the market. This increase toxicological data (principles of toxicolo-
Sub-Acute / Sub-chronic toxicity
is associated with the possibility that new gical evaluation, study protocols, toxico-
(28-90 days in the diet)
and unapproved additives will appear on logical section reporting data, review of
Reproductive toxicity studies
the market, and that "old additives" will be results and conclusion, etc.). In the USA
Teratology studies in 2 species
used in unacceptable amounts. With the FDA provides the correspondentguid-
Short-term and in vitro tests for
several hundred food additives on the mar- ance of safety evaluationin their so-called
Mutagenicity / Clastogenicity
ket, it is important that each of them is "Red book" .
Chronic toxicity
subject to a thorough scientific evaluation The requirements for the toxicological
(long-term dietary, e.g. 2 year rat)
and regulation (6,7). There are funda- programme (Figure 1) are in many ways
Carcinogenicity, normally in
mental demands for approval of a food similar to those for drugs. However, since
2 rodent species
additive. The food additive has to be of additives are used in foods that may be
Mechanistic / Toxicological studies
value for the consumer or necessary for consumed by humans of all ages and
with structural analogs
the handling and distribution of the food, sometimes during the whole life time,
as well as being toxicologically accept- adverse effects are not acceptable. The
able, according to regulatory require- situation is somewhat different for drugs,
ments within the European Community. since benefits of drug use are considered
Figure 2. The value of different types of studies in
risk assessment. The value of different types of All approved food additives on the Swe- in relation to risksladverse effects. The
studies in risk assessment is increased the closer dish market have been evaluated for safety present guidelines from 1992 are pre-
we get to humans. either by Joint FAOIWHO Expert Com- sently under re-evaluation and a new edi-
mittee on Food Additives (JECFA) (8) or tion will be available early next year (9).
+ Epidemiology the Scientific Committe for Food (SCF), Animal studies are needed for the safety
which both propose ADI-values (The assessment (Figure 2), and exposure stu-
+ Animal experiments Acceptable Daily Intake). In the USA, the dies covering selected segments of the
Food and Drug Administration (FDA) total lifetime cannot replace lifetime stu-
+ Short term tests E
does the corresponding work. The JECFA dies (10). The major difference in the new
evaluations are continuously published as guidelines will be a more pronounced
monographs, whereas the earlier work of focus on ADME studies (Absorption-
SCF was not officiallypublished in detail. Distribution-Metabolism-Excretion), in-
However, SCF-reports (opinions) from cluding species comparisons of phar-
recent years are available on the Internet. macokinetics and pharmacodynamics.
difficult this area is regarding monitoring, There will alsocbe more emphasis on the
safety assessment and regulations. Thus, the most important work regarding
evaluation of use, intake and safety of food use of a more scientifically oriented case
additives in Sweden is today performed by by case approach in designing the toxi-
Food additives are specified cological programme, such as the in-
the SCF and their Working Group for Food
on the food package Additives. This Working Group consists clusion of new techniques in molecular
All food additives that are used in a food- of about 15 independent scientists, half of biology, new studies of neurotoxicology,
stuff should be declared on the label on the them are members of the SCF and the irnrnunotoxicologyand genetic toxicology.
package to make it possible for the con- other half are ad hoc experts. The author
sumer to identify them. This is called the (NGI) participates as ad hoc expert.
Purity and specifications
specification of the ingredient content. o f a-food additive
Each additive should bk declared by its When a company/producer would like There are high purity requirements for
function,for example aspartame should be to put a new additive on the EU market, food additives and the safety evaluation
labelled as a sweetener, followed by the they have to submit an application to the should include an evaluation of the
name aspartame or the specific identity SCF. In this expert group additives are production procedure and a safety evalua-
number (E-number), that is E951 for evaluated for safety and individual ADI- tion of any contaminants (6). Thus, the
aspartame. The E-number shows that the values are set before the additives are process by which the raw materials are
additive has been evaluated for safety and allowed to be used in the different food converted to the finished product should
is permitted for use within the European products (Figure 1). Furthermore, the be described. The maximum residue le-
Union (see section below). In some cases maximum amount of the additive that is vels of processing chemicals and impuri-
other information should also be given on allowed to be added to each food category ties should by the petitioner be specifiedin
the label. For example, aspartame is a is strictly regulated. The application sub- the technical description of the applica-
source of phenylalanine and this must be mitted by the petitioner, to use a new food tion. The justification for the additive, as
stated on the label since some people have additive, has to full fill certain extensive well as the reaction and fate of the additive
the disease phenylketonuria (PKU) that requirements. In order to help the appli- in food, should be described. In addition,
make them unable to metabolise this cant, the Food Additives Working Group when the production procedure is changed
amino acid. Although flavouring agents has issued guidelines which describe what for a previously permitted additive a new
are by definition additives, at present they should be included in the application and application from the petitioner is always
do not have to be individually specified or these have to be strictly followed by the needed. In future, SCF will also have the
declared. Thus, flavouring agents are de- applicant. According to the guidelines full responsibilityfor evaluating the safety
clared as a group which means that the "Presentation of an application for assess- and applicability of chemicals intended to
individual substances as well as the total ment of a food additiveprior to its authori- be used in specific production procedures
number of added substancesare unknown. zation" fiom 1992 (6) a lot of information of food additives.
Food additives. Use, intake and safety

Setting the Acceptable Daily applied. Thus, a flexible approach, using What happens when the additive
Intake (ADZ-value) additional factors, is adopted in order to intake exceeds the ADZ?
The quality/value of the data that can be perform risk assessments. Different food additives will get different
used in risk assessmentvaries accordingto ADI-values and a relatively low AD1 in-
the set of data obtained from epidemio- "Old" food additives dicate that adverse effects may have
logy, animal experiments and short term are re-evaluated occurred at lower doses in the animal stu-
tests. When a food additive is already on There is also an ongoing process of re- dies. For most additives a numerical ADI-
the market it is very difficult to detect evaluation of the use, intake and safety of value (mg/kg bwlday) has been deter-
adverse effects. That is the reason for the food additives that have previously been mined that shows how much it is possible
extensive requirements for safety testing approved and are already on the market. to eat each day during the whole life
before a new food additive is permitted to New toxicological and epidemiological without any risks of adverse effects. As
be used on the market. data may appear or there may be suspicion previously shown, the normally used AD1
From the results in the toxicological file that the food consumption pattern has is based on animal data and is calculated
submitted by the petitioner, the SCF changed and thus the intake of certain food with a safety factor of at least 100. How-
calculates the ADI-value for each additive additives has increased. This work of re- ever, if there are no adverse effects, it is
(Figure 1). For the safety evaluation of evaluation is performed by scientific possible to use the additive according to
pesticides and contaminants an "acute groups, regulatory agencies, Nordic pro- the principles of Good Manufacturing
reference dose" is used to detect levels ject groups, SCF and JECFA. The docu- Practice (quantum satis), i.e. the minimum
where toxic effects appear after a single mentation of all food additives on the amount necessary to produce the desired
dose. This "acute reference dose" has been market is presently under evaluation by effects.
used as a bench mark for a short-termAD1 the Nordic Group on Food Toxicology and Obviously AD1 does not represent an
(11). Such a dose and evaluation is not Risk Assessment (NNT). The intention is absolute threshold for toxicity. If an
applicable for food additives since the to find additives that need further evalua- individual or a group of individuals ex-
safety should cover daily intake during the tion or additives where the documentation ceed the ADI, this simply means that there
whole life time, i.e. the acceptable daily is deficient in important respects. These is a reduced margin or safety factor bet-
intake (ADI). In 1987 the WHO defined selected additives could then be re-evalu- ween the intake by that individual and the
the AD1 as "An estimate by JECFA of the ated by SCF. no effect level in animals (18). Excursions
amount of food additive, expressed on a However, for many of the "old" food above the AD1 are probably not frequent,
body weight basis, that can be ingested additives, we have good historical know- but it has to be realised that it may be
daily over a lifetime without appreciable ledge of their safety and use, which means difficult to prevent extreme consumers
health risk" (7). The "safe" dietary level or that there are only a limited number of with bizarre food habits from exceeding
daily intake for animals is known as the food additives that may be of concern the AD1 (19). Consequently,it is not hazar-
"no observable effect level", NOEL. The today. dous to health when the additive intake
AD1 is calculated from the NOEL as occasionally exceeds the ADI, although
follow: Are children more individuals should not exceed the AD1 for
prolonged periods. The significance of a
NOEL sensitive than adults? higher intake has to be assessed on an
AD1 (mg/kg bwlday) = The calculated AD1 is valid for both
Safety factor additive by additive basis and depends on
children and adults (14). Today there is a the magnitude and duration of the excess
Man may be more sensitive than the most lot of discussion whether children are intake (18). Nevertheless, excursions of
sensitive species tested in the toxico- more sensitive than adults, but no scienti- food additive intake above the AD1 should
logical studies, or certain specific indivi- fic evidence has really shown this to be always trigger additional considerations
duals and subgroups may show a greater true. In a workshop, organized by ILSI and actions (20).
sensitivity than the general population. Europe (15), focused on the applicability
Uncertainty or safety factors are used to of the AD1 to infants and children it was
extrapolate from a group of animals to an strongly recommended that special safety Possible interactions
average human and from average humans factors for children should not be used between food additives
to potentially sensitive groups (12). To (14,16). Consequently, special ADIs should The amounts of the different food addi-
allow for these possibilities a large "un- not be established and the overall AD1 tives that may be added to different food
certainty factor" is applied to the NOEL, should cover all sensitive segments of the stuffs are strictly regulated. However, for
normally 100, i.e. a safety factor of 10 is population, irrespective of age. some foodstuffs it is possible to use a
applied for intra-species differences and Children, however, have a higher food number of different food additives, some-
another factor of 10for interspecies differ- intake than adults, on aper kg body weight times with similar chemical characteri-
ences. This means that the average daily basis, and they also have other dietary stics and purpose. This has raised concern
intake (ADI) will not exceed one-hund- habits and food preferences compared that structurally related, and even un-
redth of that which has been shown to be with adults (17). In addition, it must be related, food additives toxicologically
without observed effects in animal tests. borne in mind that only a very limited could interact and have synergistic effects,
The need for an extra safety factor depend- number of food additives are allowed to be possibly due to a potential to share com-
ing on the nature of toxicity have been used in food specially intended for small mon sites and mechanisms of toxic actions
debated (13). However, this safety factor children. or common pathways of elimination. How-
can be reduced when pharmacokinetic Thus, a different exposure pattern to ever, studies have shown that many of
and/or phannacodynamic data are avail- food additives needs to be kept in mind those additives which can not be excluded
able. When human data are available it is when they are used in products pre- from having such synergistic effects had
possible to accept a safety factor of 10, ferentially consumed by children. This low intakes or even are alternatives for the
whereas when animal data are less may also have a precautionary impact on same application (21). Furthermore, from
convincing a safety factor of 1000 may be the risk assessmentof certain food additives. these results it was concluded that syn-
Ilback and Busk

Table 2. Sweetening capacity of artificial sweeteners in relation to saccharose, including ADI-values

indicate whether a specific food additive
and a calculationof how much saccharose (in terms of sweetness)that has to be consumed to reach the
in these products may be over-consumed
specific ADI.
or if there is no evident risk that the AD1 is
Sweetener ADI-value Times sweeter than Amount (g) of saccharose
exceeded. Using such calculations, food
mgkg bwlday saccharose equivalent to the AD1
additives of high priority can be identified
(in terms of sweetness) and monitored. It is also possible that this
monitoring could partly be performed in
co-operation with the producers, since
Acesulphame K 0 - 15 130 - 200
Aspartame 0 - 40 approx. 200 they have to document the amount of each
Cyclamate 0 - 11") 25 - 30 food additive used for each specific food
Saccharin 0-5 approx. 300 product.
When it was checked with the producers
*) Opinion and suggestion to reduce the AD1 of cyclamate to 0 - 7 mgkg bwlday has during the in Sweden, the added amount of, for
fall of 2000 been presented by the SCF. example, sweeteners in light soft drinks
was mostly below the maximum per-
mitted concentrations. For example, the
added amount of aspartame is usually 270-
ergistic effects between additives are less ent food additives. For example, intake of 500 mg/l (maximum level 600 m g ) , sacch-
likely to occur and does not represent a azo colours may cause an increased fre- arin 12-30 r n g (maximum level 80 mg/l)
health concern. quency of hypersensitivity reactions among and acesulfame K 50-2 10mg/l (maximum
sensitive individuals (23), whereas high level 350 mg/l). These data are compar-
Interactions -food additives and intake of sweeteners among diabetics and able to the results obtained in a Norwegian
other environmental factors others who consume large amounts of intake study on sweeteners, a project
We are constantly exposed to nutrients, artificially sweetened food may pose an within the Nordic co-operative work (26),
food additives and environmental pollu- increased risk that AD1 is exceeded (24). showing that light soft drinks had lower
tants that may cause toxicity and/or have In general, a low AD1 indicates that ad- than maximum concentrations of the
effects on the immune system (22). We are verse effects have occurred at relatively sweeteners.Thus, it can be concluded that,
also constantly invaded by a variety of low doses in animal studies. However, for at least for sweeteners, the producers seem
micro-organisms, but mostly we do not the safety assessment of human use it is to comply with the regulations.
develop clinical disease. Thus, since there important to include also the biologicalcu-
may be synergisticeffects and since we are lating and chemical characteristics of the Whichfood additives
concomitantly exposed to micro-orga- food additive. The different ADI-values need to be monitored?
nisms and compounds that at specific for four artificial sweeteners can be used There are several types of information that
levels of intake may induce toxicity, it is to exemplify the complexity of use and have to be available before a human risk
important that interdisciplinary consi- safety of food additives (Table 2). The AD1 assessment of food addditive intake can be
derations are included and used in all for the sweeteners is almost comparable, performed (Figure 3). Information about
future risk-assessment. This is, however, but there is a large difference in their the toxic dose and adverse reactions can be
not a specific issue for food additives, but practical use as well as in their stability, obtained from human and animal studies
should always also be kept in mind when taste and sweetening capacity (25). For and data from alternative methods as
possible interactions are considered in the example, saccharin is ten times as sweet as previously described. This forms the basis
process of safety evaluation of food cyclamate.Thus, 10times more cyclamate for calculating the AD1 and a safe intake.
additives. has to be added to get the same sweetness, However, it is also important that the use
which means that there is clearly a higher of food additives is toxicologically accep-
risk to exceed the AD1with cyclamate than table. Depending on the level of AD1 and
How do we apply the ADI with saccharin (23). In addition, the AD1 in which foodstuffs and at what levels the
in risk management? for cyclamate was recently re-evaluted by
According to the previously described the SCF and it was suggested that the AD1 additive is going to be used, it is important
safety evaluation different food additives should be decreased to 7 mg/kg bw/day. to calculate the total food additive intake
will be assigned different ADI-values. This means that the total intake may have of the individual normal consumer in
There are different risk groups for differ- to be further decreased. This low sweet- order to avoid exceeding the ADI. Thus,
information about the actual use of food
ness combined with high estimated intake additives and intake of food that contains
Figure 3. Risk assessment of food additives. Im- is the reason why, in order to keep cycla-
portant information on toxicology, food con- mate intake below the ADI, Sweden pre- the specific additive is critical. This in-
sumption and concentration of the additive in formation can be obtained from food addi-
food that are needed for a relevant risk evalua- viously prohibited the use of cyclamate in tive content and/or food intake monitoring
tion. products other than table top sweeteners. programmes. In addition, in the Swedish
food regulations from the NFA ("Kun-
What is the toxic dose and the ADZ-value? What levels of additives gorelse om livsmedelstillsatser") the
+ Human data are actually used? maximum permitted amount of each addi-
+ Animal data Sales data from GfK (Gesellschaft fiir tive is specified for each group of food-
+ Alternative methods Konsumentmarkt und absatzforschung),a stuffs (27).
What is the intake of the additiveporn food? company that registers the total quantity of With several hundred food additives on
+ Calculation/estimation of intake different food products sold by individual the market, many of them widely used in
- food data + chemical analysis dealers, can be used to estimate the total different kinds of food products, it is
- consumer surveys consumption of a certain food product. impossible to monitor the intake and use
- biological monitoring Using these data, a potential high con- of all additives. A SCOOP (Scientific
sumption can be identified and also Cooperation between the European Union
Food additives. Use, intake and safety

and Norway) project called "Develop- exceeded (34). Possible study methods amount present in that individuals diet.
ment of methodologies for the monitoring that can be used for estimation of the The parent compound, metabolites or
of food additive intake across the Euro- intake of food additives are the following: other biological markers of exposure can
pean Union" was completed recently (28). Per capita estimates: It requires an esti- be used in these studies.
In the project report a restricted strategy mate of the total amount of the food addi- The budget method: This was originally
for monitoring food additive use and tive entering the food supply in a given developed in Denmark as a simple, inex-
intake was presented. It is only possible to country or a region. This amount divided pensive way of assessing proposed maxi-
perform and focus monitoring on a by the number of consumers gives a figure mum use levels of food additives to ensure
randomised selection of food additives for the average intake. It is a simple cal- that the ADIs for the additives are not
and products. Depending on the size of culation, but it gives no information on the exceeded. The method is designed to
AD1 and in which foodstuff the additive is distribution of intake within the popu- cover the worst case scenario and, there-
going to be used, it is important in risk lation. The method is used for monitoring fore, exaggerates potential additive in-
assessment to calculate the normal con- trends in additive usage. For example, it is take. The method is applicable for assess-
sumer intake in order to prevent the con- possible to determine whether the use of ing potential maximum intake in countries
sumers from exceeding the ADI. an additive is increasing or decreasing by with different eating habits. The budget
The most important food additives to repeating the usage survey at regular inter- method is, however, unsuitable for pre-
monitor have to be identified and available vals. dicting what consumers are actually in-
resources focused on those. The SCOOP By using sales data from G f l (Gesell- gesting. Where more precise information
strategy concludes that many food addi- schaft fur Konsumentmarkt und absatz- on intake is considered necessary, more
tives may be used in unrestricted amounts forschung) it is possible to estimate the precise calculations should be performed.
(quantum satis) and many are used in total consumption of a certain food pro- The budget method is valid internationally
small amounts and only in a few food duct. With these data available, a potential as it is based on the fact that there is a
products (28). Thus, efforts should be high consumption can be identified and physiological upper limit to the amount of
focused on those food additives that are also indicate whether the AD1 of a specific food and drink, and thus of food additives,
used in many different food products, in food additive is exceeded. that can be consumed each day. A further
food products that are consumed in high Duplicate diet studies: A selected group assumption is that only a certain propor-
amounts, or when there is an evident risk of individuals is asked to retain duplicate tion of the diet is likely to contain food
that some groups or individuals may portions of all the food they consume over additives. Using appropriate estimates for
exceed the ADI. Here we have certain risk the study period. These samples are aggre- the upper limits of food and drink, which
groups, such as children, diabetics, ethnic gated and analysed for the substance(s) of can be consumed each day, the budget
groups, etc. By using the SCOOP strategy interest. This method requires a high method allows regulators to allocate the
resources could be focused on additives degree of co-operation and considerable AD1 of a food additive between food and
that may cause concern. Thus, only a few effort from the participants. Realistically beverages and also to set maximum levels
dozen of the several hundred food this study can only be undertaken over of use.
additives on the market may need a short periods. It is especially useful for New methods under development: Nor-
complete monitoring. There are several studying specific population groups assu- mally the methods calculate possible
different ways to identify and estimate the med to have a high intake of the sub- intakes and do not take count of the pro-
intake of additives, but this is difficult, stance(~)of interest. bability of the intakes actually occurring.
expensive and time consuming and could Calculations using data on food con- Thus, there is a future need for new and
be performed in several different ways sumption and additive occurrence: The refined methods. Stochastic modelling is a
(28-3 1). most common approach to estimate the method in which there is a growing
intake of food additives is to combine the interest by agencies (personal communi-
Methodology of food information on the occurrence of the cation J. Lambe). The US Environmental
additive intake additive and the consumption of the spe- Protection Agency (EPA) has stated that
Which method that is the most suitable cific food that contains the additive. Addi- probabilistic analysis techniques can be
depends on the available information and tive occurrence can be obtained from se- viable statistical tools for analysing
the type of estimates that are required. veral sources, such as maximum permitted variability and uncertainty in risk assess-
Since intake is not constant for each levels, technological levels, manufactures ments.
individual or between individuals in a data and analytical data. Food consump-
population, the distribution of intakes tion information can be obtained from Alarm in the media
across the population is likely to be trade data (including market research), Even though food additives are well re-
different if it is measured over a short national diet studies at the level of the gulated both regarding intake and risks
period compared with the same measure- household, national diet studies at the in- there is a never ending debate in the media
ment made over a longer time period. dividual level, whether by 24 hour recall, how dangerous food additives are. This is
Furthermore, the consumption is often diary record, dietary history or food fre- especially true for the artificial ones.
more concentrated to certain individuals quency questionnaires, as well as from There are several examples, but in this
rather than spread generally over the total other more limited dietary surveys con- article we will focus mainly on the arti-
population. The capacity of food con- centrating on population subgroups, geo- ficial sweetener aspartame (E95 1) and
sumption surveys can be enhanced by com- graphic regions and/or specific foods. briefly on glutamate (E620-625), which
bination with a qualitative food frequency Biological marker of exposure: Saccha- both have caused a lot of debate during
questionnaire (32). When the distribution rin is, for example, not metabolised in the many years.
of individual food intakes is available the body and is excreted unchanged in the Glutamate: The concern regarding glu-
estimation of extreme levels of intake is urine. It is therefore possible to estimate tamate has developed from people visiting
possible, but with precautions (33). When the intake by measuring the amount pre- the Far East reporting that they have ex-
using techniques of worst case analysis sent in urine over a 24-hour period, since perienced a kind of hypersensitivity reac-
there is an evident risk that the AD1 is it will approximately be equal to the tion after eating food containing glutamate.
Ilback and Busk

However, there is no scientific evidence in is still no indication of any adverse effect food additive will never be approved for
the literature that glutamate can cause associated with aspartame consumption. use by the SCF or reach the market. In the
these effects (35). Glutamic acid com- Recently, some publications have scientific literature it has also been con-
prises some 20% of dietary protein and the indicated that the eating behaviour is cluded from both epidemiological data
food additive use does not markedly in- changed by aspartame (43,44). It has even and risk assessment of experimental data
crease the total body burden. The explana- been speculated by Erlanson-Albertsson that virtually all foodborne carcinogenic
tion for the reported reaction may be the that aspartame specifically increases the risk is attributed to naturally occurring
normal use of a variety of strong spices in hunger and appetite (45,46). In the media carcinogens in the diet and not to food
these food cultures. Thus, based on this author has even extended this additives (48).
present toxicological and epidemiological speculation to involve todays problem
studies there is no scientific reasons that with body weight increases and fatness of Additives in the future
invalidate the appropriateness of allocat- people in the western world. However, There is a continuous development of new
ing an AD1not specified to glutamate (36). this hypothesis has no scientific back- and more efficient food additives. At the
Aspartame: Of all the 300 food addi- ground and the results obtained in these same time regulatory requirements are
tives on todays market aspartame is the studies are probably a normal response to increased. In the future a new international
most investigated additive and it has been a decreased food energy intake. Aspar- agency within the EU will be responsible
on the market for almost 20 years (37). tame, as other non-caloric sweeteners, is for the safety evaluation of food additives.
Even though aspartame (composed of the intended to be used as a caloric-reducing This mean that more resources can be
three naturally constituents phenylala- agent and when the energy intake is focused on the safety evaluation of old and
nine, aspartic acid and methanol) is meta- reduced a normal physiological response new food additives. Several new additives
bolised in the body as other proteins and would be to feel hunger. There are no are in the pipe-line to be marketed within
generates a small amount of energy (17 kJ/g) animal studies, even with life-time ex- the near future. One example is a new
it is regarded as non-caloric (38-40). In the posure to high doses of aspartame, or sweetener with a sweetness 500-600 times
media it has repeatedly been stated that human studies that show any tendency to that of saccharose, which for consumer
aspartame may cause an array of adverse weight increase. Thus, there is no scien- safety means that much less will be con-
reactions, such as headache, difficulties to tific basis to support the hypothesis that an sumed of a sweetener with a higher safety.
remember, neurological problems, excema, increased food intake or fatness would Another important step ahead will be the
heart problems, weight changes, halluci- specifically be induced by the sweetener risk evaluation of processing agents and
nations, difficulties to breath, hyperacti- aspartame. increased requirements in the additive
vity, mental retardness in children, and purity specifications of very low residue
brain tumours. Nothing of this is true (41). Risk management and levels of contaminants.
These rumours developed as a result of a actions taken by NFA
clinical intake study of aspartame per- When alarm and debate appear in the me- Conclusions
formed by the Center for Disease Control dia it is important for NFA to evaluate all The use of food additives is, on the basis of
(CDC) in USA (37). In this study CDC possible scientific documentation and to the ADI, strictly regulated within EU. The
asked questions to find clinical effects, but contact other European agencies for enforced regulation should provide assur-
no adverse reactions were reported. Today information and discussion of possible ance that the AD1 will not be exceeded
these questions have been turned into risks associated with the consumption of when foods are ingested. The overall AD1
actual findings of adverse reactions. Never- the specific additive. It is fundamental to should cover all segments of the popula-
theless, after this study that showed aspar- realise that a "0" risk does not exist and it tion, irrespective of age. AD1 does not re-
tame to be safe, the use as a sweetener was is impossible to avoid all risks for certain present a threshold for toxicity and exceed-
by FDAin 1983extended to all food stuffs. individuals (47). The decisions concern- ing the AD1 simply means a reduced safety
The reason why people still are con- ing the acceptable level of risk and the size factor. Thus, it is not hazardous to health
cerned about aspartame is partly due to an of the population at risk are political choices. when the additive intake occasionally
article by Olney et al. (42) in which the It is important for NFA to participate in exceeds the ADI, although individuals
authors have misused data on CNS (brain) international scientific expert organisa- should not exceed the AD1 for prolonged
tumour incidence during 75-92 from tions that perform safety work regarding periods.
SEER (41), the cancer register in USA. use and intake of food additives. This infor- Most information indicates that food
Olney is right in that there was a small mation is then in various ways communi- additives are used well within accepted
increase in tumour incidence around the cated to the consumers and media, which limits. However, there is a very special
time for introduction of aspartame, but it is is especially important today when a lot of group of food additives, the flavouring
scientifically really doubtful whether an data circulate on Internet that is not scien- agents, that needs to be further evaluated
increased cancer incidence would appear tifically based or have been critically eva- for safety and regulated similarly to other
at the same time as a substance is in- luated. For this purpose NFA uses its web food additives.
troduced on the market. Nevertheless, site and its own journal V4r Foda. Before There are some food additives where
during the following years this increase in it is reasonable and possible for NFA to consumption can be assumed to be high in
brain tumours flattened out even though act, scientifically based data are needed. If relation to the AD1 and those have to be
there was a tremendous increase in aspar- such data are available, actions will al- identified. Of the approximately 300 food
tame consumption. Finally, during the last ways immediately be taken. additives on the market there are only a
years there has been a decrease in brain There is, for example, a lot of concern in few that needs to be more thoroughly
tumours that was not even reported by the media and among the consumers that checked. For such food additives it is
Olney et a1 (42). After this report several food additives may induce cancer. This is important that monitoring of use and in-
countries in Europe, as well as the scien- an adverse reaction that never would be take is performed, that potentially sen-
tific expert organisations SCF and FDA accepted by any agency. If there are only sitive groups and/or individuals are iden-
have independently re-evaluated previous indications in the safety evaluation of a tified and that intake calculations are
and new data. After all this new work there potential risk for carcinogenic effects the performed to ensure that the AD1 is not
Food additives. Use, intake and safety

exceeded. Excursions of food additive Reg Toxicol Pharmacol2000;30:42-7. Corti A (Ed.). World Rev Nutr Diet. Basel,
Hemnan JL: Acute intake assessment: evolu- Karger. 1999; vol85: pp. 178-200.
intake above the AD1 should always trig- tion within the Codex Committee on pesticide Bergsten C, Andersen NL, Barbieri HE, Faer-
ger additional considerations and actions. residues. Fodd Add Contaminants 2000; 17: den K, Hallikainen A, Ilback N-G, Jorgensen K,
All additives except for the flavouring 551-5. Lund-Larsen K: Guidelines for intake calcula-
Renwick AG: Toxicokinetics in infants and tions of food additives and contaminants.
agents should be declared on the label of children in relation to the AD1 and TDI. Food Report from a Nordic Expert Group, pp. 1-82,
the package by its function and also by the Add Contamin 1998;15:17-35. 2000 In press
name or the specific identity E-numbers. Renwick AG: The use of an additional safety or Lambe J, Kearney J, Leclercrq C, et al.:
uncertainty factor for nature of toxicity in the Enhancing the capacity of food consumption
Thus, it should be possible for the con- estimation of acceptable daily intake and toler- surveys of short duration to estimate long term
sumer to identify all food additives on the able daily intake values. Regul Toxicol Phar- consumer-only intakes by combination with a
macol l995;22:250-61. qualitative food frequency questionnaire. Food
package of the foodstuff. The food addi- Larsen JC, Pascal G: Workshop on the appli- Add Contaminants 2000; 17:177-87.
tive has to be of value for the consumer or cability of the AD1 to infants and children. Food Chambolle M: Assessment of extreme levels of
necessary for the handling and distribu- Add Contaminants 1998;15:1-10. chronic food intakes. Reg Toxicol Pharmacol
Corti A (ed.):Intake studies - Lessons Learnt. 2000;30: 13-8.
tion of the food, as well as being toxico- Proceedings 1995 ISA Annual Conference, pp. Hallas-Moller T: Results of a first screening-
logically acceptable according to Euro- 1-156, November 27,1995, Brussels, Belgium. using the budget method. Intake-studies -
Lawrie CA: Different dietary patterns in Lessons Learnt. Proc 1995 ISA Symposium, Ed
pean Community regulatory require- relation to age and the consequences for intake A. Corti (Brussels: International Sweeteners
ments. The NFA has no special recom- of food chemicals. Food Add Contamin 1998; Assocciation), pp. 57-77.
mendations regarding the consumption of 15:75-82. Stevensson DD: Monosodium glutamate and
Renwick AG, Dorne JL, Walton K: An analysis astma. Review. J Nutr 1999;130:1067-73.
food additives, except that the ADI, as of the need for an additional uncertainty factor Walker R: The significance of excursions above
determined for each specific additive by for infants and children. Reg Toxicol Pharmacol the ADI. Case study: Monosodium glutamate.
the SCF, should not be exceeded. 2000;31:286-96. Reg Toxicol Pharmacol2000;30: 119-21.
Renwick AG, Walker R: An analysis of the risk Ilback N-G, Beckman-Sundh U, Busk L:
of exceeding the acceptable or tolerable daily Sotningsmedlet aspartam. Vad 2.1-ratt och fel i
ACKNOWLEDGEMENTS intake. Regul Toxicol Phannacol1993;18:463-80. debatten? VAr Foda 1997;3:24-26. In Swedish
The valuable comments of Professor Stuart Slorach Larsen JC, Richold M: Report of workshop on Renwick AG: Clinical pharmacology-The
(NFA) are greatly appreciated. the significance of excursions of intake above metabolism of intense sweeteners. Xenobiotica
the ADI. Reg Toxicol Pharmacol2000;30:2-12. l986;16: 1057-71.
REFERENCES Speijers GJA: Precision of estimates of an AD1 Joint FAO/WHO Expert Committee on Food
Fennema OR: Food additives - an unending (or TDI or PTWI). Reg Toxicol Pharmacol Additives (JECFA): Aspartame In. Toxicolo-
controversy. Am J Clin Nutr 1987;46:201-3. 2000;30:87-93. gical evaluation of certain Food Additives.
Zinck 0 , Hallas-Moller T (eds): E-nummer Groten JP, Butler W, Feron VJ, Kozianowski G, WHO Food Additives Series 1982;16: 28-32.
bogen. Tillsaetningsstoffer i maden: Virkningar Renwick AG, Walker R: An analysis of the Karim A, Burns T: Metabolism and pharma-
og bivirkninger. Aschehoug Dansk Forlag A/S, possibility for health implications of joint cokinetics of radiolabelled aspartame in normal
Denmark, 1999 pp. 1-286, (ISBN 87-11- actions and interactions between food additi- subjects. In: The Clinical Evaluation of a Food
12754-6). In Danish ves. Reg Toxicol Pharmacol2000;3 1:77-91. Additive - Assessment of Aspartame. Tschanz
Council of Europe. Natural sources of fla- Ilback N-G: The need to include assessment of C, Butchko HH, Stargel WW, Kotsonis FN
vourings - Report No. 1. Council of Europe synergistic effects, bioavailability, malnutri- (Eds.). CRC Press, New York, 1996, pp.67-86.
Publishing, Germany, 2000 pp. 1-276, (ISBN tion, etc, in risk evaluation. Nordic Seminar on Levy PS, Hedeker D: Letter to the Editor. J
92-87 1-4324-2). Toxicological and Microbiological Risk Neuropathol Exp Neuorol 1996;55:1280.
Ilback N-G, Beckman-Sundh U, Busk L: Cya- evaluation of Food, October 7-9, 1998, Koge Olney JW, Farber NB, Spitznagel E, Robins
nider. Ett livsfarligt gift - men smA mhgder Denmark. pp. 19-22,2001.In press LN: Increasing brain tumor rates: Is there a link
behover vi. VAr Foda 1997;4:16-20. In Swedish Thuvander A: Hypersensitivity to azo colouring to aspartame? J Neuropathol Exp Neurol
Munro IC, Kennepohl E, Kroes R: A procedure agents. Tartrazine in food may cause rash and 1996;55:1115-23.
for the safety evaluation of flavouring sub- asthma. Laartidn 1995;92:296-8. In Swedish Lavin JH, French SJ, Read NW: The effect of
stances. Joint FAO/WHO expert committee on Ilback N-G., Alzin M, Jahrl S, Enghardt H, sucrose- and aspartame-sweetened drinks on
Food Additives. Fd Chem Toxic 1999;37:207-32. Busk L: Sweetener intake and exposure study energy intake, hunger and food choice of
Commission of the European Communities. among diabetics in Sweden. SLV-rapport 2: 1- female, moderately restrained eaters. Internat J
Food Sciences and Techniques. Reports of the 50,2000 (ISSN 1104-7089). In Swedish Obesity 1997;21:37-42.
Scientific Committee on Food: Presentation of Widenfalk A, Bergsten C, Ilback N-G: Sot- Melanson KJ, Westerterp-Plantega MS, Camp-
an application for assessment of a food additive ningsmedel istallet for socker. VAr Foda 1998;7: field LA, Saris WHM: Blood glucose and
prior to its authorization. 1992 (ISBN 92-826- 12-17. In Swedish patterns in time-blinded males, after aspartame,
0135-8). Bergsten C: Inntak av kunstige sotstoffer: Ace- carbohydrate and fat consumption, in relation to
Walker R: Toxicity testing and derivation of the sulfam-K, aspartam, cyklamat og sakkarin. sweetener perception. Br J Nutr 1999;82:437-66.
ADI. Food Add Contamin 1998;15:11-6. Statens naeringsmiddeltilsyn (SNT) 1998; Erlanson-Albertsson C: Aspartam - ett sot-
Joint FAO/WHO Expert Committee on Food report 4: 1-89, (ISSN 0802-1627). In Norwegian ningsmedel med ovantade effekter. Scand J
Additives (JECFA): Toxicological evaluation Livsmedelsverkets kungorelse om livsmedels- Nutr/N&ingsforskning 2000;44:82-3. In Swedish
of certain Food Additives. Published as mono- tillsatser. SLV FS 1999:22. In Swedish Erlansson-Albertsson C: Aptitreglering och
graphs in WHO Food Additives Series 1973- SCOOP Report. Development of methodolo- makron&ingsamnen. Perspektiv - Tidskrift om
2000. gies for the monitoring of food additive intake socker och n&ing, published by Danisco Sugar
Commission of the European Communities. across the European Union, Task 4.2 - Final 2000;2:5-8. In Swedish
Food Sciences and Techniques. Reports of the Report. SCOOP/INT/ REPORT/2,1998. Verger P, Gamier-Sagne I, Leblanc J-C: Iden-
Scientific Committee on Food: Guidance on Gibney MJ, Lambde J: Estimation of food tification of risk groups for intake of food che-
submissions for food additive evaluations by additive intake: methodology overview. Food micals. Reg Toxicol Pharmaco12000;30: 103-8.
the Scientific Committee on Food. 2001 In press Add Contaminants 1996;l3:4O5- 10. Scheuplin RJ: Perspectives on toxicological
Knudsen I: Temporal equivalaence between test Renwick AG: Intake of intense sweeteners. In: risk - An example: Food borne carcinogenic
species and humans: General toxicity issues. Low-calorie sweeteners. Present and future. risk. Crit Rev, Food Sci Nutr 1992;32:105-21.

Scand J Nutr/N&ingsforskning4/00