Content uploaded by Sebastien La Vieille
Author content
All content in this area was uploaded by Sebastien La Vieille on Jan 04, 2019
Content may be subject to copyright.
Special Article
Dietary restrictions for people with glucose-6-phosphate
dehydrogenase deficiency
S
ebastien La Vieille, David E. Lefebvre, Ahmad Firas Khalid, Matthew R. Decan, and Samuel Godefroy
Glucose-6-phosphate dehydrogenase (G6PD) deficiency is the most common en-
zyme deficiency in the world and renders those affected susceptible to potentially
severe oxidative hemolysis. Although the resulting hemolysis is most often associ-
ated with drug exposure, it has also been reported after consumption of certain
foods. With the exception of review articles that reiterated the historical knowledge
that fava beans can provoke severe oxidative hemolysis in individuals with G6PD
deficiency, very few articles have examined the safety of other food ingredients and
food additives for people with G6PD deficiency. Some articles that associated spe-
cific foods with hemolysis appeared to be speculative and based on limited infor-
mation. The objective of this review was to examine the association between foods,
including food additives, and the triggering factors of acute hemolysis. The litera-
ture was searched for studies and case reports on food consumption and G6PD de-
ficiency. In this review, fava beans were found to be the only food for which there
is conclusive clinical evidence linking the risk of hemolytic anemia to individuals
with G6PD deficiency. Food additives, at their permitted level of use in North
America, can be consumed safely by most patients with G6PD deficiency.
INTRODUCTION
Glucose-6-phosphate dehydrogenase (G6PD) deficiency
is the most prevalent enzyme deficiency in the world,
affecting at least 330 million individuals worldwide.
1
This metabolic enzyme plays an important role in pro-
tecting erythrocytes against oxidative stress, thereby
preventing hemolysis. Individuals with G6PD defi-
ciency are susceptible to oxidants such as drugs (eg,
some antimalarial drugs or antibiotics) or chemicals
(eg, naphthalene in mothballs).
2
Fava beans are also his-
torically known to provoke severe incidents of acute he-
molytic anemia in individuals with G6PD deficiency.
3
G6PD deficiency is an X-linked, hereditary genetic de-
fect arising from mutations in the G6PD gene. The
genetic deficit is carried by females but primarily affects
males (90% of people with G6PD deficiency are males).
It largely affects populations of African, Mediterranean
Basin, Indian, Southeast Asian, or Latin American
descent.
1
Several reviews have focused on drugs to be
avoided by individuals with G6PD deficiency.
4–6
Apart
from recommending the avoidance of fava beans,
3
how-
ever, very few published articles have assessed informa-
tion on other foods as triggering factors of acute
hemolysis. Here, a brief overview of G6PD deficiency is
provided, followed by an evidence-based review that
was conducted to identify food ingredients, food addi-
tives, and contaminants that should be avoided by indi-
viduals with G6PD deficiency.
Affiliation: S. La Vieille, D.E. Lefebvre, and M.R. Decan are with the Bureau of Chemical Safety, Food Directorate, Health Canada, Ottawa,
Ontario, Canada. S. La Vieille and S. Godefroy are with the Department of Food Sciences, Faculty of Agriculture and Food Sciences,
Universit
e Laval, Qu
ebec City, Qu
ebec, Canada. A.F. Khalid is with the Health Policy PhD Program, McMaster University, Hamilton, Ontario,
Canada. S. Godefroy is with the Institute of Nutrition and Functional Foods, Universit
e Laval, Qu
ebec City, Qu
ebec, Canada.
Correspondence: S. La Vieille, Bureau of Chemical Safety, Food Directorate, Health Canada, 251 Sir Frederick Banting Driveway, Ottawa,
ON K1A 0K9, Canada. Phone: þ1-343-542-3889; Email: sebastien.lavieille@canada.ca.
Key words: G6PD deficiency, favism, food additive, food coloring agent, herbal product.
V
CThe Author(s) 2018. Published by Oxford University Press on behalf of the International Life Sciences Institute.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
doi: 10.1093/nutrit/nuy053
Nutrition Reviews
V
R
Vol. 0(0):1–11 1
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
Role of G6PD enzyme in red blood cell metabolism
and integrity
In the various types of cells in the body, several antioxi-
dant systems use the reduced form of nicotinamide ade-
nine dinucleotide phosphate for its reducing capacity.
The G6PD enzyme catalyses the first step of the pentose
phosphate pathway by converting glucose-6-phosphate
to glucose-6-phosphogluconolactone and reducing the
cofactor nicotinamide adenine dinucleotide phosphate
(NADP) to nicotinamide adenine dinucleotide phos-
phate hydrogen (NADPH). Production of NADPH is
essential to the protection of the cells from oxidative
stress.
2
Moreover, because erythrocytes generate only
NADPH with this pathway, they are more susceptible
than other cells to destruction from oxidative stress.
7
The G6PD enzyme is also necessary to regenerate the
reduced form of glutathione that is produced with 1
molecule of NADPH. Thus, G6PD deficiency leads to a
depletion of glutathione, which is essential for the re-
duction of hydrogen peroxide and oxygen radicals and
the maintenance of hemoglobin and other red blood
cell proteins in the reduced state.
4
G6PD deficiency as a genetic disease
To date, more than 400 different biochemical variants
of G6PD have been described on the basis of their di-
verse biochemical characteristics, and a total of 217 dif-
ferent point mutations have been identified in the
G6PD gene.
8,9
Most of these biochemical variants result
in phenotypes that are asymptomatic until exposed to
oxidative triggers. Exposure to these triggers most com-
monly manifests in 1 of 2 clinical outcomes.
10
In the
first situation, which corresponds to the majority of de-
ficiencies, the mutation causes only mild impairment of
the enzyme’s function, which is not sufficient to result
in symptoms. Therefore, affected individuals are often
unaware of their status. In the less-common situation,
the mutation greatly reduces the activity of the G6PD
enzyme, which can lead to severe clinical manifesta-
tions, including hemolytic anemia and, in newborns,
neonatal jaundice due to increased bilirubin levels.
Chronic hemolytic anemia and frequent acute hemo-
lytic anemia can be associated with potentially fatal
acute renal failure, which can occur at any age.
The biochemical variants of the G6PD enzyme are
categorized into 5 classes on the basis of enzymatic ac-
tivity in red blood cells and clinical manifestations.
4,7
(1) Class I variants are associated with a severe deficit of
enzymatic activity, which leads to chronic nonsphero-
cytic hemolytic anemia. These variants occur sporadi-
cally at a very low frequency in any part of the world.
Patients with chronic nonspherocytic hemolytic anemia
have less than 10% of normal G6PD enzyme activity. In
some patients with class I variants, the residual levels of
enzyme activity are so low that a sufficient concentra-
tion of NADPH cannot be maintained, leading to
chronic hemolytic anemia with recurrent episodes of
acute hemolysis. These patients often have a history of
neonatal jaundice and have the most clinically severe
manifestations and may be dependent on transfusions.
8
Olusanya et al
11
reported that G6PD deficiency is a
common risk factor for severe neonatal hyperbilirubi-
nemia in low- and middle-income countries, such as
India and Pakistan. (2) Patients with class II variants
also have less than 10% of the normal G6PD activity but
without chronic nonspherocytic hemolytic anemia.
Class II variants include the common Mediterranean
(G6PD Mediterranean) and common severe Asian var-
iants (eg, G6PD Canton). Patients with these variants
suffer only intermittent acute hemolysis typically caused
by infection, exposure to oxidative drugs, or consump-
tion of fava beans.
4,7,12
(3) Class III variants are moder-
ately deficient (patients have 10%–60% residual enzyme
activity) and include the common African (G6PD A
)
form and Mahidol variants in Southeast Asia. Patients
with class III variants have intermittent hemolysis gen-
erally triggered by infection, oxidative drugs, foods.
4,7,12
(4) Class IV variants have a very mild or no enzyme de-
ficiency (patients have >60% of normal enzyme activ-
ity) and are not associated with hemolysis.
4,7,12
(5) Class
V variants are associated with higher-than-normal en-
zyme activity.
4,7,12
The most common variants are the G6PD
Mediterranean variants (class II), which affect popula-
tions from the Mediterranean Basin, the Middle East,
and the Indian subcontinent, and the G6PD A
variant
(class III), which accounts for the vast majority of
G6PD deficiency in Africa. In Asia, a number of poly-
morphic variants are present, each with a unique distri-
bution throughout the continent. However, common
variants in Asia are the G6PD Canton variant (class II)
and the Mahidol variant (class III). The G6PD Canton
variant is notably frequent in Singapore and Malaysia
but is also present in China. The Mahidol variant is typ-
ically present in Thailand but is also found throughout
Southeast Asia.
12–14
Screening tests for G6PD deficiency are available,
but routine screening of newborns is not commonly
performed in the United States and Canada. The tests
are usually used only to identify causes of newborn
jaundice in hospitals in North America. Introducing
screening tests may be a useful preventive approach be-
cause G6PD deficiency can lead to an increased risk
and earlier onset of hyperbilirubinemia, which may re-
quire treatment by phototherapy or exchange
transfusion.
13
2Nutrition Reviews
V
R
Vol. 0(0):1–11
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
Acute hemolytic anemia is the most common man-
ifestation of G6PD deficiency and occurs when red
blood cells are under oxidative stress, which is typically
triggered by infections, oxidant drugs, or food.
Associations with hepatitis viruses A and B, cytomega-
lovirus, pneumonia, and typhoid fever have been
reported.
4
The severity of hemolysis can be affected by
many factors, including concomitant drug administra-
tion, liver function, and age. However, the exact mecha-
nism by which the infections can trigger hemolysis is
unknown.
8
Oxidant drugs reported as triggers include certain
antibiotics (dapsone, nitrofurantoin), antimalarial drugs
(primaquine), and some analgesics (phenazopyridine).
6
The consumption of fava beans is a commonly
reported trigger of hemolytic anemia. The potential im-
plication of other food ingredients triggering hemolytic
anemia is the major topic of this review.
Prevalence of G6PD deficiency in North America
The prevalence of G6PD deficiency in North America is
not well known. The samples in most studies conducted
to date are not representative of the general population
and/or have been small in size. However, a large retro-
spective US study published in 2006 identified the pres-
ence and degree of G6PD deficiency in military
personnel by sex and self-reported ethnicity.
15
In a sam-
ple of more than 63 300 participants, the reported prev-
alence was 2.37%, with most individuals having only
moderate enzyme deficiency (class III variants). Rates
of G6PD deficiency were highest in African American
males (12.2%) and females (4.1%) and in Asian males
(4.3%). Most males were found to have class III var-
iants, while most females had class IV variants and did
not report symptomatic hemolysis. According to 2014
data published by the US Census Bureau, 12.2% of
African American men, 4.3% of Asian American men,
and 2% of Hispanic American men, together represent-
ing approximately 3.2 million American men, could be
possibly affected by symptomatic G6PD deficiency.
There is no study related to the prevalence of
G6PD deficiency in Canada. According to Statistics
Canada, Canada’s population was estimated at
34 460 065 in 2016. African Canadians, West Central
Asian and Middle East Canadians, Asian (Eastern,
Southeast, and Southern Asians) Canadians, and
Hispanic Canadians represent 2%, 1.8%, 9.6%, and
1.9% of the total Canadian population, respectively. On
the basis of US estimates
15
of G6PD deficiency in males,
at least 150 000 men might be affected by symptomatic
G6PD deficiency in Canada.
16
Knowing that Caucasian individuals can also be af-
fected by G6PD deficiency, and, taking into account
recent immigration, these numbers are probably an un-
derestimate of the number of people affected in North
America. These rough estimates indicate that, while
G6PD deficiency is a monitored medical problem in
countries of the Mediterranean Basin, Africa, the
Middle East, and Asia, it also represents a possible
growing health concern in other parts of the world, as
pointed out in several reports.
13,17,18
Literature search and classification
A literature search was conducted for studies and case
reports on food consumption and G6PD deficiency
published until June 2017. The search was conducted in
Scopus and PubMed databases, including MEDLINE,
Embase, Compendex, and Cochrane database. The fol-
lowing search terms were used: “glucose 6 phosphate
dehydrogenase deficiency” and “food(s)” and “adverse
event(s) or side effect(s) or h(a)emolysis”. The keywords
“food supplement(s),” “dietary supplement(s),” “food
ingredient(s),” “chemical(s),” “dye(s),” and “additive(s),”
were used. In addition, “faba,” “bean,” and “favism,”
were used along with the keywords “food(s) sup-
plements” and “food(s) ingredients.” A total of 376
articles, including review articles, were identified using
this search procedure. After duplicates were removed,
titles and abstracts of articles were screened to identify
potentially relevant articles. Ninety-seven full-text manu-
scripts, written in English or French, were finally identi-
fied as relevant and were included in this review. Details
of the clinical studies, oral exposure, laboratory findings,
and the authors’ conclusions, as well as information pro-
vided in case reports, were analyzed.
Similar to the approach used in other publications
on drugs and supplements,
6,19,20
foods and food deriv-
atives were classified into 1 of the 3 following groups
(Table 1): (1) foods and food derivatives that should be
avoided in patients with G6PD deficiency; (2) foods
and food derivatives that can be safely consumed at
certain dosages by patients with G6PD deficiency; and
(3) foods and food derivatives cited in the literature
but for which there is insufficient evidence to preclude
their consumption by patients with G6PD deficiency.
Also taken into consideration in this classification
were the influence of G6PD variants on the degree of
food-induced hemolysis and the amount of food
consumed.
Foods that should be avoided by patients with G6PD
deficiency
Fava bean (Vicia faba var major) and vetch (Vicia
sativa). Favism is a term describing hemolytic anemia
that results from the ingestion of fava beans (Vicia faba
Nutrition Reviews
V
R
Vol. 0(0):1–11 3
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
var major). Its association with G6PD deficiency is well
documented in the scientific and medical literature.
Several authors have already reviewed the epidemiol-
ogy,
3,12,21
pathophysiological mechanisms,
4,22–25
and
clinical features of favism.
3,4
Therefore, fava beans have
already been validated as a food that should be avoided
in patients with G6PD deficiency.
Known since antiquity,
12,26
acute hemolysis caused
by fava bean ingestion in patients with G6PD deficiency
presents strong analogies to hemolysis caused by oxida-
tive drugs.
27
Two pyrimidine glycosides (vicine and
convicine) isolated from fava beans have been impli-
cated as favism-inducing factors. There are various ana-
lytical methods and bioassays for their quantification in
fava beans.
28
The active substances within these pyrimi-
dines are their aglycone derivatives.
22,26,29
Divicine is
the aglycone derivative of vicine, and isouramil is the
aglycone of convicine. Both of these oxidizing substan-
ces rapidly overwhelm the already diminished
glutathione-generating capacity of G6PD-deficient cells
and also have direct effects on red blood cell function.
In vitro studies have shown that divicine reduces the ac-
tivity of catalase, an enzyme that, like glutathione, con-
tributes to hydrogen peroxide removal and requires
NADPH for maintenance of normal activity.
30
Similarly, isouramil treatment of erythrocytes in vitro
caused a marked decrease in cellular membrane
deformability, which was presumed to be a factor in
erythrocyte sequestration in individuals with G6PD de-
ficiency.
31
Acute intravascular hemolysis (ie, loss of
membrane integrity of erythrocytes in the blood
circulation) is the most common form of favism.
Individuals who develop favism are almost invariably
deficient in G6PD, but, owing to the polymorphism of
the deficiency, not all persons with the deficiency are
sensitive to fava beans.
12
Thus, anemia is usually most
severe and most common with the class I variant and is
somewhat less so with the Mediterranean variant (class
II).
21,32,33
Favism has been observed occasionally in
patients with the G6PD A variant (class III).
5,14,34,35
It is
considered rare among people from Thailand or
Southeast Asia, possibly because of different G6PD
mutations or a different consumption pattern of fava
beans among these populations.
32,36
Whatever the re-
gion, favism is generally more frequent in children (be-
tween the ages of 1 and 5) but can occur in adults.
14,37
It may also occur on one eating occasion but not on an-
other,
37,38
possibly because of the amount of fava beans
ingested in the meal, the type of beans prepared (raw,
fresh, cooked, or dried), or the concentration of pyrimi-
dine glycosides in the fava bean cultivars.
39
The species Vicia faba belongs to the genus Vicia.
Within that species, Vicia faba var major is the botani-
cal variety (subspecies), commonly known as fava beans
or broad beans. Other varieties within the species Vicia
faba with seeds reported to be involved in favism are
Vicia faba var minor and Vicia faba var equina.
40
These
2 varieties, which have seeds that are smaller in size, are
grown mainly for use as protein supplements in animal
feed. The main feature of Vicia faba var major is the
large size of its seeds, intended for human consump-
tion. Fresh and raw seeds are harvested at the immature
Table 1 Classification of foods and food derivatives for consumption by patients with glucose-6-dehydrogenase (G6PD)
deficiency
Foods and food derivatives that
should be avoided in patients with
G6PD deficiency
(2) foods and food derivatives that can be safely
consumed at certain dosages by patients with
G6PD deficiency
Foods and food derivatives for which
there is insufficient evidence to preclude
their consumption by patients with G6PD
deficiency
Fava bean (Vicia faba var major) Azo-containing food coloring agents at permitted
levels of use in foods:
•In Canada: amaranth, Allura Red, Indigotine,
Sunset Yellow FCF, Tartrazine, Citrus Red No. 2
and Ponceau SX
•In the United States: Orange B, Allura Red,
Indigotine, Sunset Yellow FCF, Tartrazine, Citrus
Red No. 2
Quinine-containing beverages (except
in breastfeeding mothers of infants
with G6PD deficiency)
Vetch (Vicia sativa) (almost never
present on the North American
market)
Rhizoma Coptidis, which is sometimes labeled as
Coptis chinensis (Chinese herbal extract present in
some dietary supplements)
Fenugreek seeds
Orange RN (not permitted as a food
additive in Canada or the United
States)
Acalypha indica (herbal extract present in some die-
tary supplements and used Ayurvedic medicine)
Pumpkin
Vitamin C (ascorbic acid) at dosages <2 g/d in
adults or <400 mg/d in young children (aged
1–3 y)
Unripe peaches
Tea extracts
Ginkgo biloba extract
4Nutrition Reviews
V
R
Vol. 0(0):1–11
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
stage, while dry seeds are harvested at the mature
stage.
40
Other species of the genus Vicia are used in agri-
culture and for animal feed, such as Vicia sativa, also
called vetch, which is grown mainly as a fodder crop
and a green fertilizer.
10
Vicia sativa is rarely consumed
by humans, but a few cases of direct consumption are
known in some Mediterranean regions, and at least 1
case of hemolytic anemia due to vetch consumption in
a child with G6PD deficiency in Turkey has been
reported.
41
Vetch and fava beans contain similar
amounts of vicine and convicine, the chemicals from
which favism-active aglycones are derived. Therefore,
the scarcity of acute hemolysis reports related to vetch
consumption in areas where G6PD deficiency is en-
demic could be a reporting oversight as a consequence
of mild symptoms (as for fava beans). In addition, vetch
is most commonly grown and consumed in poor rural
areas, where emergency departments are not generally
accessible.
42
The content of vicine and convicine in fava beans
varies with the variety of bean, the growing conditions,
and the stage at which the seeds are harvested. Fresh
fava bean seeds are the edible parts of the plants with
the highest content of vicine and convicine. The
amount of these pyrimidine glycosides in dried fava
beans is approximately 50% lower than that in fresh
fava beans. Additionally, fresh seeds also contain b-glu-
cosidases and ascorbate, which could play a role in the
severity of symptoms, as outlined below in the section
Vitamin C (ascorbic acid).
43
The peak seasonal inci-
dence of favism in the Mediterranean Basin coincides
with the harvesting of the bean.
21
Hence, raw and fresh
fava beans are considered more likely to cause fav-
ism,
4,21
while dry bean seeds and dry bean flour, used
specially as food ingredients, pose a lower risk.
10
There are few data available to enable either the as-
sessment of the effects of temperature on fava beans or
the consequences of long-term storage of fava beans
prior to consumption. b-glucosidase and ascorbate are
reputedly sensitive to cooking, whereas vicine and con-
vicine are considered relatively resistant to cooking and
to germination treatments.
10
In North America, fava beans are sold fresh, dried,
frozen, and canned. Fresh fava beans usually begin to
appear in markets in late May to June. Frozen and dried
beans are available year round. There are not enough
responses in North American consumption surveys to
provide an estimate of fava bean consumption, but
some articles related to the consumption of pulses (ie,
dry beans, peas, and lentils) are available. Thus, the
main sources of pulses in the Canadian diet in 2012
were mung beans, kidney beans, baked beans, bean
soups, chili, and pulse-containing Mexican or Hispanic
mixed dishes.
44
On any given day, 13.1% of Canadians
consume pulses, with the Asian population having the
highest consumption. Data from the 2009 US National
Health and Nutrition Examination Survey (NHANES)
show that only 7.9% of adults in the United States con-
sume dry beans and peas.
45
However, the NHANES
analysis did not include mung beans and other sprouted
forms of beans that were included in the Canadian
study. When those were removed from the Canadian
analysis, the proportion of pulse consumers dropped
from 13.1% to 10.7%.
44
Those studies were based on
consumption data collected in the mid 2000s, and levels
of consumption in North America may have increased
as a result of recent population migrations from
Southeast Asia and the Middle East.
Foods, food derivatives, and herbal products that can
be safely consumed at certain dosages by patients
with G6PD deficiency
Synthetic food coloring agents. Several review articles
proposed that food additives consisting of azo dyes
(also known as aniline dyes or acid dyes) may be associ-
ated with hemolysis in individuals with G6PD defi-
ciency.
19,46
However, accurate descriptions of these
hemolysis incidents after ingestion were missing in the
literature.
The permitted synthetic food coloring agents in
Canada are amaranth, Allura Red, Brilliant Blue FCF,
Erythrosine, Fast Green FCF, Indigotine, Sunset Yellow
FCF, Tartrazine, Citrus Red No. 2, and Ponceau SX;
however, the last 2 are limited to very specific uses
(Citrus Red No. 2 in skins of whole oranges, and
Ponceau SX in fruit peels and maraschino cherries).
47
Except for amaranth and Ponceau SX,
48
which are not
permitted for use in foods in the United States, these
same synthetic food coloring agents, as well as Orange
B, are also permitted in the United States as food color-
ing additives subject to certification.
49
Of these syn-
thetic food coloring agents, amaranth, Allura Red,
Indigotine, Sunset Yellow FCF, Tartrazine, Citrus Red
No. 2, Ponceau SX, and Orange B contain either an azo
or aniline moiety.
The literature was searched for case reports of ane-
mia associated with exposure to synthetic food coloring
agents. Notably, azo dyes, because they sometimes con-
tain trace levels of subsidiary dyes or impurities such as
aniline, were cited as a potentially important consider-
ation for individuals with G6PD deficiency.
19,46
Indeed, when the intermediates of azo dye synthe-
sis, 2-chloroaniline and 3-chloroaniline, were adminis-
tered experimentally in large quantities (10–160 mg/kg/
d) by gavage in rats for 13 weeks, they induced a reduc-
tion in red blood cell counts and in the hemoglobin
Nutrition Reviews
V
R
Vol. 0(0):1–11 5
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
content of red blood cells.
50
The reduced red blood cell
count was likely due to hemolytic anemia rather than to
hematopoietic stem cell toxicity, because the pathology
showed that hematopoietic tissues and kidneys had in-
creased microscopic markers of red blood cell produc-
tion alongside lesions that are typical of red blood cell
destruction. In rats, aniline hydrochloride administered
at doses between 10 and 100 mg/kg of body weight (ap-
proximately equivalent to 250 and 2500 mg of aniline
per kilogram, respectively) for 1 to 4 weeks produced
hemoglobin adducts.
51
However, the Canadian Food and Drug
Regulations
47
and the US Code of Federal Regulations
49
require food coloring agents to meet specifications that
limit the levels of impurities, including aniline. When
synthetic food coloring agents are used in food, the
trace amount of aniline that may be present is diluted
to the order of micrograms per kilogram (ppb) of ani-
line in the overall diet, because the coloring agents are
used in small amounts. Therefore, these levels of con-
taminants are not expected to pose a hazard to human
health.
Naphthalene can also be detected as an impurity at
very low levels in some food coloring agents.
52
This
contaminant is commonly used as a moth repellent in
the form of balls or disks. Several cases of hemolytic
incidents following exposure to naphthalene mothballs
have been reported in the literature, notably in children
with G6PD deficiency. These incidents were reported
following exposure to naphthalene vapors
53
or follow-
ing ingestion of naphthalene-containing anointing
oil.
54,55
No incident following exposure to naphthalene
as an impurity of food coloring agents has been
reported in the literature.
Orange RN, an azo dye that is no longer permitted
in foods, is also found as an impurity in Sunset Yellow
FCF at a level of no more than 1% in Western coun-
tries.
56
Several cases of hemolysis in African individuals
with G6PD deficiency were linked to the consumption
of barbecued meat (red suya) that contained the pure
coloring agent Orange RN.
57–59
Those reports lacked
details confirming the chemical composition and cau-
sality; however, studies in which pigs and rats were fed
red suya demonstrated reduced red blood cell counts
following exposure.
60,61
Regardless, Orange RN is not a
permitted food additive in Canada or the United States.
Overall, these animal data suggest that aniline or
naphthalene could have a negative impact on erythro-
cyte metabolism and could cause damage if encoun-
tered at high doses.
10
However, in the context of the
possible presence of either compound as an impurity in
synthetic coloring agents, levels would be extremely
low. As mentioned above in this section, regulations
limit the levels of these compounds that are permitted
as impurities in coloring agents, and coloring agents are
used at relatively low concentrations that further dilute
the presence of either of these impurities in food.
Additionally, with regard to the permitted coloring
chemicals themselves, oral exposure of pigs to amaranth
at a dose of 500 to 750 mg/kg/d or to Allura Red at 1000
to 1500 mg/kg/d for 76 days did not impact the erythro-
cyte count or the hemoglobin concentration.
62
For
comparison, similar synthetic azo dyes that are not cur-
rently permitted as additives in food in Canada were
also reviewed. This data may aid in developing read-
across toxicity assessments. Recently, Health Canada
performed a screening assessment of the azo dye New
Coccine (Ponceau 4R), which is used in some other
countries.
63
When mice were administered 1625 mg/kg/
d—a dose above that considered by Health Canada to
be the lowest-observed-adverse-effect level—in a 19-
month study, there was statistically significant mild ane-
mia in the first 6 months of the study. When consider-
ing approaches to extrapolate other data from
toxicology studies in healthy wild-type rodents, it is
conceivable that a decrease or increase in the G6PD en-
zyme level following chemical exposure might reflect
impairment or an adaptive coping response to oxidative
stress. Therefore, an increase or decrease in activity
might serve as a marker of chemical risk to humans
lacking G6PD enzyme. For example, healthy rats
injected intraperitoneally with a fava bean extract and
5 mg of diethyl maleate for 15 days had decreased serum
G6PD.
64
The synthetic azo solvent dye Methyl Yellow,
also known as Butter Yellow, is no longer permitted in
food. Dietary administration of Methyl Yellow for 1 or
7 months in rats induced an increase in liver G6PD ac-
tivity at those time points. However, this may have been
secondary to hepatic hypertrophy and was not observed
at the 3-month time point.
65,66
Oral gavage of rats with
the azo dye amaranth at 340 mg/kg/d for 14 days did
not alter the activity of G6PD in liver or kidney.
67
Similarly, rats gavaged for 9 days with amaranth at
85 mg/kg/d did not show modified liver G6PD activity,
but administration of amaranth at 115 mg/kg/d in com-
bination with vitamin A reduced G6PD activity.
68
Evidence of the oxidative capacity of the chemicals
is also useful, since oxidative capacity is a known mech-
anism by which divicine and isouramil in fava beans
induces favism. A conceivable additional mechanism in
individuals with G6PD deficiency could be an impact of
the reduced enzyme activity on degradation of the dye.
For example, in media containing rat cecal contents, the
addition of cofactors that included the enzyme G6PD
was associated with increased metabolism of ama-
ranth.
69
Thus, in individuals with G6PD deficiency, the
potential longer exposure to the parent amaranth mole-
cule may be a factor because the parent molecule is
6Nutrition Reviews
V
R
Vol. 0(0):1–11
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
more cytotoxic to leukocytes than the metabolite naph-
thionic acid.
70
Mechanistic data comparing the cytotox-
icity of amaranth with that of its metabolites for
erythrocytes are lacking.
While the potential risk of permitted synthetic food
coloring agents in individuals with G6PD deficiency
merits assessment, there are no reports that conclusively
implicate these agents in cases of anemia in humans.
Although some reviews suggested that aniline dyes
should not be consumed by individuals with G6PD de-
ficiency,
19,46
there is no evidence that consumption of
the permitted synthetic food coloring agents at their
permitted level of use presents a risk to individuals with
G6PD deficiency.
Chinese herbal products. Chinese herbal products can be
consumed as dietary supplements, and the increasing
consumption of these products in North America
emphasizes the need to address the possible complica-
tions they may cause in patients with G6PD deficiency.
However, the potential toxicity of these products in
individuals with G6PD deficiency is controversial, and
the number of documented cases of hemolysis due to
the use of unspecified Chinese herbal products is scarce
in published literature. Neither the purity nor the actual
composition of these products is well documented,
which complicates the evaluation of any potential
hazard.
The Department of Health of the Government of
Hong Kong Special Administrative Region issued in
2016 an alert card
71
advising individuals with G6PD de-
ficiency to avoid exposure to Chinese herbal medicines,
including Rhizoma Coptidis,Flos Lonicerae,Flos
Chimonanthi, and Calculus Bovis. However, there has
been very little documented evidence in the medical lit-
erature associating these herbs with hemolysis in indi-
viduals with G6PD deficiency. In 1 case from
Singapore, severe hemolysis in a newborn with G6PD
deficiency who had been fed a berberine-containing
Chinese herbal medicine was reported.
72
Berberine is
an alkaloid from the rhizomes of Coptis chinensis
Franch (family Ranunculaceae), officially recognized in
the Chinese Pharmacopoeia as Rhizoma Coptidis (also
known as Huang Lian) and frequently found in tradi-
tional Chinese herbal formulas.
73
A second report is re-
lated to the administration of Coptis chinensis in an
infant with G6PD deficiency in 1996.
74
In 2008, Ko
et al
75
investigated the oxidative effect of 18 commonly
used Chinese herbal medicines on human G6PD-
deficient red blood cells. In this study, the oxidative ac-
tion of 6 of the herbal medicines (Rhizoma Coptidis,
Cortex Moutan,Radix Rehmanniae,Rhizoma Polygoni
Cuspidati,Radix Bupleuri, and Flos Chimonanthi) was
demonstrated by both a reduction in the activity of
glutathione and an increase of methemoglobin in red
blood cells. However, a review by Valaes
76
in 1994
found no convincing epidemiological data relating
Chinese herbs to hemolysis in G6PD deficiency except
for the cases reported in Singapore by Wong
72
in 1980
and by Yeo and Tan
74
in 1996. In addition, the products
used in both case reports were home-based prepara-
tions, and the ingredients in these preparations were
not fully identified. Ho et al
77
similarly concluded that
studies exploring the toxicity of Rhizoma Coptidis were
contradictory and that several of them had only estab-
lished the toxicity of berberine through intraperitoneal
and intravenous injections. Therefore, they considered
oral administration of berberine to be clinically safe at
the recommended dosage (15 mg/kg/d).
77
According to
Fok,
78
although it is a common assertion in the Chinese
community that a number of herbs can cause hemolysis
in babies with G6PD deficiency, some of the herbs de-
scribed in the Hong Kong alert card are, paradoxically,
still consumed by pregnant women and often used for
the treatment of some neonatal ailments, including
jaundice. In 2013, Singapore’s Health Sciences
Authority lifted the ban for selling Chinese herbs con-
taining berberine, as there are no major safety concerns
when berberine and its associated Chinese herbs such
as Rhizoma Coptidis are used appropriately. However,
the Health Sciences Authority also recommended label-
ing requirements to warn against berberine use in indi-
viduals with G6PD deficiency.
79
The increased popularity of dietary supplements
containing Chinese herbs has led to a need for further
research to clarify the potential relationship between
these products and G6PD deficiency. However, on the
basis of information available in the literature and, no-
tably, the very low number of reports of hemolysis
caused by the use of unspecified Chinese herbal prod-
ucts in individuals with G6PD deficiency, the products
containing these herbs, when ingested in a reasonable
manner, can probably be consumed safely by most per-
sons with G6PD deficiency. Nonetheless, caution
should be recommended for individuals with G6PD de-
ficiency when Rhizoma Coptidis or Coptis chinensis is in
the list of ingredients of Chinese herbal products.
Other herbal products. In a systematic review published
in 2016, Lee et al
20
indicated that Acalypha indica,an
Asian plant notably used in traditional Indian
(Ayurvedic) medicine, could possibly be hazardous for
individuals with G6PD deficiency, since cases of acute
hemolysis after consumption of Acalypha indica by
individuals with G6PD deficiency had been reported.
Dietary supplements containing extracts of Acalypha
indica can easily be bought on the Internet. Little is
known about the toxicity of this plant in individuals
Nutrition Reviews
V
R
Vol. 0(0):1–11 7
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
with G6PD deficiency, and, as indicated by Lee et al,
20
the dose and purity of the extracts used were not indi-
cated in reported cases. Although mechanistic evidence
is lacking, several case reports of hemolysis have been
published in journals that are not indexed in conven-
tional databases. Therefore, individuals with G6PD defi-
ciency should be advised to use caution when
consuming Acalypha indica.
Vitamin C (ascorbic acid). Vitamin C, generally known
to have antioxidant properties, has unexpectedly been
cited in several reports as triggering hemolysis when
used at high doses in individuals with G6PD defi-
ciency.
80–83
Besides its ability to provide reducing
equivalents for a variety of biochemical reactions,
vitamin C functions physiologically as a water-soluble
antioxidant. It readily scavenges reactive oxygen and ni-
trogen species as well as hypochlorite. The one- and
two-electron oxidation products are easily regenerated
by glutathione and NADPH.
84
In G6PD deficiency, the
pentose phosphate pathway reduces NADP to NADPH
less efficiently and leads to a depletion of glutathione,
thereby increasing free radicals and oxidative stress,
which, in red blood cells, leads to hemolysis. Although
vitamin C is a part of the antioxidant defense system,
incubation of erythrocytes with a solution of 0.2mM
vitamin C induced oxidative stress with depletion of
glutathione but increased flux through the pentose
phosphate pathway.
85
Similarly, in vitro experiments
with both normal and G6PD-deficient erythrocytes
showed that a solution of 5mM ascorbate alone or in
combination with divicine from fava bean increased
production of hydroxyl radical markers, which was fur-
ther enhanced by the addition of chelated iron EDTA.
86
Ascorbic acid in vitro at a solution of 1 to 9 mM also de-
creased the amount of reduced glutathione antioxidant
in erythrocytes from patients with G6PD deficiency in a
dose-dependent manner.
87
According to the present
bibliographic analysis, several cases of hemolysis in-
duced by vitamin C have been reported in patients with
a G6PD deficiency.
80–83
However, these cases occurred
after very high doses of vitamin C, either administered
intravenously
81,83
or ingested orally (3–4 g), as reported
in children who consumed a significant quantity of soft
drink in the span of 4 to 6 hours.
82
Dietary supplements typically contain vitamin C in
the form of ascorbic acid, which has bioavailability
equivalent to that of naturally occurring ascorbic acid
in foods such as orange juice. Use of vitamin C-contain-
ing supplements is relatively common and adds to the
total daily intake from food and beverages.
88
In the
United States, NHANES data (1999–2000) indicate that
approximately 35% of adults take multivitamin supple-
ments (which typically contain vitamin C), and 12%
take a separate vitamin C supplement.
89
Approximately
29% of children in the United States take some form of
dietary supplement that contains vitamin C.
90
There are
no Canadian data available related to the consumption
of dietary supplements in the general population.
The Food and Nutrition Board of the Institute of
Medicine (IOM) in the United States established a
Tolerable Upper Intake Level (UL; ie, maximum daily
intake unlikely to cause adverse health effects) for
vitamin C, which applies to both food and supplement
intakes.
91
The UL for adults is 2 g, and UL values for
children range between 400 mg (for ages 1–3 y) and
1800 mg (for ages 14–18 y). In Canada, the ULs are the
same.
92
Long-term intakes of vitamin C above the UL
may increase the risk of adverse health effects. Knowing
that the ULs do not apply to individuals receiving
vitamin C for medical treatment, and considering that
an adverse reaction was documented at an oral intake
of 3 to 4 g, it would be prudent to screen at-risk people
for G6PD deficiency before administering a high oral
dose (>2 g per day) of ascorbic acid.
Given the scientific evidence available, it does not
seem possible to set a threshold for the vitamin C dose
that would pose a risk to individuals with G6PD defi-
ciency. There is currently no concrete evidence that
oral intake of vitamin C lower than the ULs (2 g in
adults) would trigger hemolysis in individuals with
G6PD deficiency.
Foods cited in the literature but for which there is
insufficient evidence to preclude their consumption
by patients with G6PD deficiency
Quinine-containing beverages. Several reviews have clas-
sified quinine and some of its derivatives (quinidine
and mefloquine) as antimalarial drugs that can be safely
given in normal therapeutic doses to patients with
G6PD deficency.
6,12,13,93
Nonetheless, acute hemolytic
episodes in individuals with G6PD deficiency have been
described with 2 other quinine-derived antimalarial
drugs. Primaquine (8-aminoquinoline) is the most
common antimalarial drug implicated in these inci-
dents.
12,93
There is solid evidence associating prima-
quine with hemolysis and G6PD deficiency, and
therefore primaquine must be avoided by individuals
with G6PD deficiency.
6
The literature implicating chlo-
roquine (4-aminoquinoline) as a causative agent of he-
molysis in individuals with G6PD deficiency is scarce,
and some authors believe this antimalarial agent can be
safely administered to patients with G6PD defi-
ciency.
6,12,13
Others, however, are more cautious, nota-
bly because this drug has oxidative properties that could
trigger a decrease in glutathione levels.
2,93
8Nutrition Reviews
V
R
Vol. 0(0):1–11
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
A number of beverages (eg, tonic water, bitter
lemon) contain quinine derivatives in the form of hy-
drochloride or sulfate salts. In the United States, the
Food and Drug Administration limits the amount of
quinine used as a flavor in carbonated beverages to
83 ppm (ie, 83 mg/L, corresponding to approximately
30 mg of quinine per can).
94
Two cases of hemolysis in
breastfed neonates with G6PD deficiency, occurring at
7 days and 8 days of life secondary to maternal con-
sumption of tonic drinks, have been reported in
France.
95
The same authors reported a third case of he-
molysis in a breastfed 5-month-old infant whose
mother had consumed tonic water, with quinine subse-
quently found in the breast milk.
95
Similarly, women
undergoing quinine-derived antimalarial drug treat-
ment can have detectable levels of quinine in their milk,
and therefore some of these treatments are contraindi-
cated during breastfeeding until the G6PD deficiency
status of the newborn is known.
96
The 3 described cases
occurred in very young infants through breast milk ex-
posure, but no documented case of hemolysis following
ingestion of quinine-containing beverages by children
or adults with G6PD deficiency has been reported in
the literature to date. On the basis of current informa-
tion available, a reaction after drinking several cans of
these beverages cannot be excluded, but a direct associ-
ation between consumption of quinine-based drinks
and the occurrence of severe hemolysis in adults and
children with G6PD deficiency is currently
speculative.
2,10
Other foods. Acute hemolysis in individuals with G6PD
deficiency following consumption of fenugreek seeds,
46
pumpkin,
97
unripe peaches,
98
tea extracts and polyphe-
nols
99
and Ginkgo biloba has been reported.
100
However, each of these food-triggered events was iden-
tified only in isolated case reports.
In 1 case report, fenugreek seeds were suspected to
have triggered hemolysis in a an individual with G6PD
deficiency.
46
Acute hepatitis and a
1
-antitrypsin defi-
ciency were excluded as causes of hemolysis, but neither
the diet nor possible exposure to drugs in this patient,
who had an 8-month history of poorly controlled diabe-
tes, was investigated, and therefore a causative relation-
ship between fenugreek and hemolysis could not be
demonstrated conclusively.
One hemolytic crisis after ingestion of pumpkin in
an infant with G6PD deficiency was reported in 2014.
97
The authors hypothesized that the pumpkin may have
been cross-contaminated by fava beans, and they
detected fava bean DNA in the pumpkin that the infant
consumed the day before jaundice onset. Therefore,
pumpkin was not directly responsible for the hemolytic
incident.
Another case report described a severe hemolytic
crisis in a child with G6PD deficiency who ate unripe
peaches.
98
The authors showed that an extract from the
unripe peach triggered oxidative stress in erythrocytes.
However, no oxidative agent(s) in the peach were spe-
cifically identified. Additionally, the child had an upper
respiratory tract infection treated with penicillin, which
could have contributed to the hemolysis. No other he-
molytic incidents induced by peaches in individuals
with G6PD deficiency have been reported in the litera-
ture, even though peaches are consumed in all regions
of the world. In this context, given there is only 1 iso-
lated report in a child with G6PD deficiency who had a
respiratory tract infection, it seems premature to add
unripe peaches to the list of hazardous foods for indi-
viduals with G6PD deficiency.
One study investigated the pro-oxidative effects of
tea and some polyphenols (epigallocatechin-3-gallate
and epigallocatechin) on G6PD-deficient erythrocytes
in vitro.
99
The tea extracts significantly decreased the
level of reduced glutathione in G6PD-deficient erythro-
cytes in a dose-dependent manner but did not alter the
level in normal erythrocytes. The authors believed it is
highly unlikely the plasma concentration of these com-
pounds would reach a harmful level in individuals with
G6PD deficiency under conditions of normal consump-
tion. Instead, they suggested that an additive effect
might occur if individuals with G6PD deficiency take
additional oxidative drugs. No case reports in the litera-
ture have described hemolysis when individuals with
G6PD deficiency consumed tea and/or polyphenols,
and, to date, involvement of tea and some polyphenols
in hemolysis in individuals with G6PD deficiency has
not been confirmed in vivo.
Lastly, there is 1 case report of a patient with G6PD
deficiency who developed acute hemolytic anemia after
having received an injection of Ginkgo biloba leaf ex-
tract.
100
Ginkgo biloba extracts are very commonly used
in dietary supplements, and no incident of hemolytic
anemia following oral exposure to Ginkgo biloba has
been published in the literature. As already indicated by
Lee et al,
20
it is unlikely that consumption of Ginkgo bi-
loba dietary supplement can lead to hemolysis in G6PD
deficiency.
CONCLUSION
In this review, fava beans were classified as the only
food available on the North American market for which
there is conclusive clinical evidence of a risk of hemo-
lytic anemia in individuals with G6PD deficiency.
Although recent literature mentions synthetic food col-
oring agents as potential risk factors, no supporting evi-
dence is provided. Thus, at their permitted level of use
Nutrition Reviews
V
R
Vol. 0(0):1–11 9
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
in North America, synthetic food coloring agents have
not been associated with risk to human health. Chinese
herbal extracts commonly present in dietary supple-
ments can probably be consumed safely by most
patients with G6PD deficiency, but caution should be
recommended if the list of ingredients includes
Rhizoma Coptidis or Coptis chinensis. There is no evi-
dence that oral intakes of vitamin C at dosages lower
than 2 g/d in adults would trigger hemolysis in individ-
uals with G6PD deficiency. For the other food ingre-
dients cited in the literature, it is not possible to
establish causative relationships between the occurrence
of hemolysis in individuals with G6PD deficiency and
exposure to these ingredients.
Acknowledgments
Author contributions. S.L.V. and S.G. conceived the
study. S.L.V., D.E.L., and A.F.K. contributed signifi-
cantly to the data interpretation and analysis. S.L.V.,
D.E.L., A.F.K., M.R.D., and S.G. participated in the
writing and critical revision of the article.
Funding/support. This study was funded in full by
Health Canada. Health Canada is the federal depart-
ment responsible for helping Canadians maintain and
improve their health.
Declaration of interest. The authors have no relevant
interests to declare.
REFERENCES
1. Nkhoma ET, Poole C, Vannappagari V, et al. The global prevalence of glucose-6-
phosphate dehydrogenase deficiency: a systematic review and meta-analysis.
Blood Cells Mol D is. 2009;42:267 –278.
2. Wajcman H, Galacteros F. Glucose 6-phosphate dehydrogenase deficiency: a pro-
tection against malaria and a risk for hemolytic accidents [in French]. C R Biol.
2004;327:711–720.
3. Luzzatto L, Arese P. Favism and glucose-6-phosphate dehydrogenase deficiency. N
Engl J Med. 2 018;378:1068–1 069.
4. Cappellini MD, Fiorelli G. Glucose-6-phosphate dehydrogenase deficiency. Lancet.
2008;371:64–74.
5. Frank JE. Diagnosis and management of G6PD deficiency. Am Fam Physician.
2005;72:1277–1282.
6. Youngster I, Arcavi L, Schechmaster R, et al. Medications and glucose-6-phosphate
dehydrogenase deficiency: an evidence-based review. Drug Saf. 2010;33:713–726.
7. World Health Organization Working Group. Glucose 6-phosphate dehydrogenase
deficiency. 1989;67:601–611.
8. Gomez-Manzo S, Marcial-Quino J, Vanoye-Carlo A, et al. Glucose-6-phosphate dehy-
drogenase: update and analysis of new mutations around the world. Int J Mol Sci.
2016;17:2069. doi:10.3390/ijms17122069
9. Minucci A, Moradkhani K, Hwang MJ, et al. Glucose-6-phosphate dehydrogenase
(G6PD) mutations database: review of the “old” and update of the new mutations.
Blood Cells Mol D is. 2012;48:154 –165.
10. Agence Franc¸aise de S
ecurit
e Sanitaire des Aliments. Avis relatif
alademande
d’
elaboration de recommandations concernant l’alimentation des personnes por-
teuses d’un d
eficit en G6PD [in French]. Referral no. 2006-SA-0033. https://www.
anses.fr/fr/system/files/NUT2006sa0033.pdf. Published August 25, 2016. Accessed
May 28, 2017.
11. Olusanya BO, OsibanjoFB, Slusher TM. Risk factors for severeneonatal hyperbiliru-
binemia in low and middle-income countries: a systematic review and meta-anal-
ysis. PloS O ne. 2015;10:e0117 229. doi:10.1371/ journal.pone.01172 29
12. Beutler E. Glucose-6-phosphate dehydrogenase deficiency: a historical perspec-
tive. Blood.2008;111:16–24.
13. Beutler E. G6PDdeficiency. Blood. 1994;84:3613–3636.
14. Mason PJ, Bautista JM, Gilsanz F. G6PD deficiency: the genotype-phenotype asso-
ciation. Blood Rev. 2007;21:267–283.
15. Chinevere TD, Murray CK, Grant E Jr, et al. Prevalence of glucose-6-phosphate de-
hydrogenase deficiency in U.S. Armypersonnel. Mil Med. 2006;171:905–907.
16. Statistics Canada. Census profile, 2016 census. 2017. https://www12.statcan.gc.ca/
census-recensement/2016/dp-pd/prof/details/page.cfm?Lang¼E&Geo1¼PR&
Code1¼01&Geo2¼&Code2¼&Data¼Count&SearchText¼Canada&SearchType¼
Begins&SearchPR¼01&B1¼All&TABID¼1 Accessed June 28, 2018.
17. Sgro M, Campbell DM, Kandasamy S, Shah V. Incidence of chronic bilirubin en-
cephalopathy in Canada, 2007–2008. Pediatrics. 2012;130:e886–e890.
18. Watchko JF, Kaplan M, Stark AR, et al. Should we screen newborns for glucose-6-
phosphate dehydrogenase deficiency in the United States? J Perinatol.
2013;33:499–504.
19. Lee SW, Chaiyakunapruk N, Lai NM.What G6PD-deficient individuals should really
avoid. Br J Clin Pharmacol. 2017;83:211–212.
20. Lee SW, Lai NM, Chaiyakunapruk N, Chong DW. Adverse effects of herbal or die-
tary supplements in G6PD deficiency: a systematic review. Br J Clin Pharmacol.
2017;83:172–179.
21. Belsey MA.The epidemiology of favism. Bull World Health Organ. 1973;48:1–13.
22. Baker MA, Bosia A, Pescarmona G, et al. Mechanism of action of divicine in a cell-
free system and in glucose-6-phosphate dehydrogenase-deficient red cells.
Toxicol Pathol. 1984;12:331–336.
23. McMillan DC, Bolchoz LJ, Jollow DJ. Favism: effect of divicine on rat erythrocyte
sulfhydryl status, hexose monophosphate shunt activity, morphology, and mem-
brane skeletal proteins. ToxicolSci. 2001;62:353–359.
24. McMillan DC, Jollow DJ. Favism: divicine hemotoxicity in the rat. Toxicol Sci.
1999;51:310–316.
25. Xiao M, Du G, Zhong G, et al. Gas chromatography/mass spectrometry-based
metabolomic profiling reveals alterations in mouse plasma and liver in response
to fava beans. PloS One. 2016;11:e0151103. doi:10.1371/journal.pone.0151103
26. Mehta A, Mason PJ, Vulliamy TJ. Glucose-6-phosphate dehydrogenase deficiency.
Clin Haem atol. 2000;13:21–38.
27. Arese P, Gallo V, Pantaleo A, Turrini F. Life and death of glucose-6-phosphate de-
hydrogenase (G6PD) deficient erythrocytes—role of redox stress and band 3
modifications. Transfus Med Hemother. 2012;39:328–334.
28. Getachew F, Vandenberg A, Smits J. A practical toxicity bioassay for vicine and
convicine levels in faba bean (Vicia faba). J Sci Food Agric. 2018;93:5105–5111.
doi:10.1002/jsfa.9049.
29. Arese P, Bosia A, Naitana A, et al. Effect of divicine and isouramil on red cell me-
tabolism in normal and G6PD-deficient (Mediterranean variant) subjects. Possible
role in the genesis of favism. Prog Clin Biol Res.1981;55:725–746.
30. Gaetani GF, Rolfo M, Arena S, et al. Active involvement of catalase during hemo-
lytic crisesof favism. Blood. 1996;88:1084–1088.
31. Chevion M, Navok T, Pfafferott C, et al. The effects of isouramil on erythrocyte me-
chanics: implications for favism. Microcirc Endothelium Lymphatics. 1984;1:295–306.
32. Laosombat V, Sattayasevana B, Chotsampancharoen T, Wongchanchailert M.
Glucose-6-phosphate dehydrogenase variants associated with favism in Thai chil-
dren. Int J Hematol. 2006;83:139–143.
33. Merdin A, Avci F, Guzelay N. Glucose-6-phosphate dehydrogenase deficiency pre-
sented with convulsion: a rare case. Hematol Rep. 2014;6:5266. doi:10.4081/
hr.2014.5266
34. Calabro V,Cascone A, MalaspinaP, Battistuzzi G. Glucose-6-phosphate dehydroge-
nase (G6PD) deficiency in southern Italy: a case of G6PD A(-) associated with fav-
ism. Haematologica. 1989;74:71–73.
35. Galiano S, Gaetani GF, Barabino A, et al. Favism in the African type of glucose-6-
phosphatedehydrogenase deficiency (A-). BMJ. 1990;300:236.
36. Kitayaporn D, Charoenlarp P, Pattaraarechachai J, Pholpoti T. G6PD deficiency and
fava bean consumption do not produce hemolysis in Thailand. Southeast Asian J
Trop Med Public Health. 1991;22:176–182.
37. Ahmed SN. Do favic patients resume fava beans ingestion later in their life, a
study for this, and a new hypothesis for favism etiology. Hematol Oncol Stem Cell
Ther. 2013;6:9–13.
38. Arese P, De Flora A. Pathophysiology of hemolysis in glucose-6-phosphate dehy-
drogenasedeficiency. Semin Hematol. 1990;27:1–40.
39. Pulkkinen M, Gautam M, Lampi A, et al. Determination of vicine and convicine
from faba bean with an optimized high-performance liquid chromatographic
method. FoodRes Int. 2015;76:168–177.
40. Duc G. Faba bean (Vicia faba L.). Field Crops Res. 1997;53:99–109.
41. Bicakci Z. A hemolysis triggerin glucose-6-phosphate dehydrogenase enzyme de-
ficiency.Vicia sativa (vetch). Saudi Med J. 2009;30:292–294.
42. Al-Mendalawi MD. A hemolysis trigger in glucose-6-phosphate dehydrogenase
enzyme deficiency. Vicia sativa (vetch) [comment]. Saudi Med J. 2009;30:974; au-
thor reply 974–975.
10 Nutrition Reviews
V
R
Vol. 0(0):1–11
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
43. Arrigoni O,De Gara L, Tommasi F, Liso R. Changes in the ascorbate system during
seed development of Vicia fabaL. Plant Physiol. 1992;99:235–238.
44. Mudryj AN,Yu N, Hartman TJ, et al. Pulse consumption in Canadian adults influen-
ces nutrient intakes.Br J Nutr. 2012;108(suppl 1):S27–S36.
45. Mitchell DC, Lawrence FR, Hartman TJ, Curran JM. Consumption of dry beans,
peas, and lentils could improve diet quality in the US population. J Am Diet Assoc.
2009;109:909–913.
46. Sadler JM Jr., Ash S, Brito-Babapulle FM. Possible fenugreek induced haemolysis
in a patient with previously unknown G6PD deficiency. BMJ Case Rep. 2009;2009.
doi:10.1136/bcr.09 .2008.089
47. Government of Canada. Food and Drug Regulations.C.R.C., c. 870, Food and Drug
Act. B.01.045. http://laws.justice.gc.ca/eng/regulations/C.R.C.,_c._870/index.html.
Updated 2017. AccessedMay 28, 2017.
48. US Food and Drug Administration. General specifications and general restrictions
for provisional color additives for use in foods, drugs, and cosmetics. Code of
Federal Regulations, Title 21, Part 81. https://www.accessdata.fda.gov/scripts/
cdrh/cfdocs/cfcfr/CFRSearch.cfm?CFRPart¼81&showFR¼1CFRPart¼81. Updated
2017. Accessed May 28, 2017.
49. US Food and Drug Administration. Listing of color additives subject to certifica-
tion. Code of Federal Regulations, Title 21, Part 74. https://www.ecfr.gov/cgi-bin/
retrieveECFR?gp¼&SID¼3463c48f55ae08efd099682901bb9500&r¼PART&
n¼pt21.1.74. Updated 2017. Accessed May 28, 2017.
50. Hejtmancik MR, Trela BA, Kurtz PJ, et al. Comparative gavage subchronic toxicity
studies of o-chloroaniline and m-chloroaniline in F344 rats and B6C3F1 mice.
Toxicol Sci. 2002;69:234–243.
51. Zwirner-Baier I, Deckart K, Jackh R, Neumann HG. Biomonitoring of aromatic
amines VI: Determination of hemoglobin adducts after feeding aniline hydrochlo-
ride in the diet of rats for 4 weeks.Arch Toxicol. 2003;77:672–677.
52. Khera KS, Munro IC. A review of the specifications and toxicity of synthetic food
colors permitted in Canada. CRC Crit Rev Toxicol. 1979;6:81–133.
53. Santucci K, Shah B. Association of naphthalene with acute hemolytic anemia.
Acad Emerg Med. 2000;7:42–47. doi:10.1111/j.1553-2712.2000.tb01889.x
54. Ostlere L, Amos R, Wass JA. Haemolytic anaemia associated with ingestion of
naphthalene-containing anointing oil. Postgrad Med J.1988;64:444–446.
55. Trevisan A, Rossi di Schio M, Pieno M. Haemolytic anaemia after oral self-giving of
naphthalene-containing oil. J Appl Toxicol. 2001;21:393–396.
56. Yamada M, Kawahara A, Nakamura M, Nakazawa H. Analysis of raw materials,
intermediates and subsidiary colours in Food Yellow No. 5 (Sunset Yellow FCF) by
LC/MS. Food Addit Contam. 2000;17:665–674.
57. Akinyanju OO, Odusote KA. Cause of red suya syndrome is food dye orange-RN.
Lancet. 1983;2:1314.
58. Akinyanju OO, Odusote KA. Red suya syndrome. Lancet. 1983;1:935.
59. Williams CK, Osotimehin BO, Ogunmola GB, Awotedu AA. Haemolytic anaemia as-
sociated with Nigerian barbecued meat (red suya). Afr J Med Med Sci.
1988;17:71–75.
60. Gaunt IF, Brantom PG, Kiss IS, et al. Short-term toxicity of orange RN in rats. Food
Cosmet Toxicol. 1971;9:619–630.
61. Hansen E, Olsen P. Peroral toxicity of Orange RN in pigs. Early haemotological
changes. In: Leonard BJ, ed. Toxicological Aspects of Food Safety: Proceedings of
the European Society of Toxicology. Berlin: Springer Verlag; 1978:313–315.
62. Sondergaard D, HansenEV, Wurtzen G. A short-term study in thepig of the effects
on the liver and on the blood of eight azo dyes.Toxicology. 1977;8:381–386.
63. Health Canada. Aromatic azo and benzidine-based substance grouping—certain
azo acid dyes. https://www.canada.ca/en/health-canada/services/chemical-sub-
stances/substance-groupings-initiative/aromatic-azo-benzidine-based.html.
PublishedJune 18, 2016. Accessed May 28, 2017.
64. Koriem KM, Arbid MS, El-Gendy NF. The protective role of anise oil in oxidative
stress and genotoxicity producedin favism. J Diet Suppl.2016;13:505–521.
65. Albrecht R, Manchon P. Influence of the hydrosoluble vitamin content of the diet
and treatment by barbiturates on several effects of azoic dyes. Ann Nutr Aliment.
1973;27:191–212.
66. Albrecht R, Manchon P, Lowy R. Influence of the protein level of the diet on some
effects of amaranth and butter yellow in the rat. Food Cosmet Toxicol.
1973;11:383–389.
67. HolmbergD. The effect of amaranth treatment on some kidney and liver enzymes
in the rat. Toxicol Appl Pharmacol. 1978;46:257–260.
68. Holmberg D. Effect of amaranth, Ponceau 4R and/or vitamin A on enzyme activi-
ties of the rat liver. FoodCosmet Toxicol. 1978;16:1–5.
69. Maris B, Verheyden L, Samyn C, et al. Influence of the composition of in-vitro azo-
reducing systems on the degradation kinetics of the model compound amaranth.
J Pharm Pharmacol. 2002;54:197–203.
70. Lefebvre DE, Ross N, Kocmarek AL, et al. In vitro immunomodulation of spleno-
cytes from DO11.10 mice by the food colouring agent amaranth. Food Chem
Toxicol. 2017;110:395–401.
71. Department of Health, Government of Hong Kong Special Administrative Region.
Glucose-6-dehydrogenase (G6PD) deficiency. http://www.dh.gov.hk/english/
main/main_cgs/files/DH2289E%20G6PD%20English.pdf. Revised 2016. Accessed
May 28, 2017.
72. Wong HB. Singapore kernicterus. Singapore Med J. 1980;21:556–567.
73. Tjong Y, Ip S, Lao L, et al. Analgesic effect of Coptis chinensis rhizomes (Coptidis
Rhizoma) extract on rat model of irritable bowel syndrome. J Ethnopharmacol.
2011;135:754–761.
74. Yeo KL, Tan CC. Severe hyperbilirubinaemia associated with Chineseherbs. A case
report. Singapore Paediatr J. 1996;38:180–182.
75. Ko CH, Li K, Ng PC, et al. Pro-oxidative effects of Chinese herbal medicine on
G6PD-deficient erythrocytes in vitro. Toxicol In Vitro. 2008;22:1222–1227.
76. Valaes T. Severe neonatal jaundice associated with glucose-6-phosphate dehydro-
genase deficiency: pathogenesis and global epidemiology. Acta Paediatr Suppl.
1994;394:58–76.
77. Ho CE, Goh YL, Zhang C. From prejudice to evidence: the case of Rhizoma
Coptidis in Singapore. Evid Based Complement Alternat Med. 2014;2014:871720.
doi:10.1155/20 14/871720.
78. Fok TF. Neonatal jaundice–traditional Chinese medicine approach. J Perinatol.
2001;21(suppl 1): S98–S100; discussion S104–S1 07.
79. Health Science Authority, Singapore Government. HSA allows Chinese proprietary
medicines containing berberine for sale in Singapore from 1 January 2013 [press
release]. Health Science Authority website. http://www.hsa.gov.sg/content/hsa/
en/News_Events/Press_Releases/2012/hsa_allows_chinese.html. Published
September22, 2012. Accessed May 28, 2017.
80. Burton C, Kaczmarski R. Oxidative haemolysis secondary to high-dose vitamins in
a woman with glucose-6-phosphate dehydrogenase deficiency. Br J Haematol.
2003;121:201.
81. Huang YC, Chang TK, Fu YC, Jan SL. C for colored urine: Acute hemolysis induced by
high-dose ascorbic acid. Clin Toxicol. 2014;52:984. doi: 10.3109/15563650.2014.954124
82. Mehta JB, Singhal SB, Mehta BC. Ascorbic-acid-induced haemolysis in G-6-PD defi-
ciency. Lancet. 1990;336:944.
83. Rees DC, Kelsey H, Richards JD. Acute haemolysis induced by high dose ascorbic
acid in glucose-6-phosphate dehydrogenasedeficiency. BMJ. 1993;306:841–842.
84. European Food Safety Authority. Scientific opinion on dietary reference values for
vitamin C (NDA panel). EFSA J. 2013;11:3418–3486.
85. Zhang ZZ, Lee EE, Sudderth J, Yue Y, et al. Glutathione depletion, pentose phos-
phate pathway activation, and hemolysis in erythrocytes protecting cancer cells
from vitamin C-induced oxidativestress. J Biol Chem. 2016;291:22861–22867.
86. De Flora A, Benatti U, Morelli A, Guida L. Oxidant damage of normal and glucose
6-phosphate dehydrogenase (G6PD)-deficient red blood cells is enhanced by
iron-EDTAcomplex. Biochem Int. 1983;7:281–290.
87. Calabrese EJ, Moore GS, McCarthy MS. The effect of ascorbic acid on sodium
nitrite-induced methemoglobin formation in glucose-6-phosphate dehydroge-
nase-deficient erythrocytes. Ecotoxicol Environ Saf. 1983;7:410–415.
88. National Institutes of Health. Vitamin C. Fact sheet for health professionals.
https://ods.od.nih.gov/factsheets/VitaminC-HealthProfessional/. Updated 2016.
Accessed May 28, 2017.
89. Radimer K, Bindewald B, Hughes J, et al. Dietary supplement use by US adults:
data from the National Health and Nutrition Examination Survey, 1999–2000. Am
J Epidemi ol. 2004;160:339–349.
90. Picciano MF, Dwyer JT, Radimer KL, et al. Dietary supplement use among infants,
children, and adolescents in the United States, 1999-2002. Arch Pediatr Adolesc
Med. 2007;161:978–985.
91. Food and Nutrition Board, Institute of Medicine. Dietary Reference Intakes for
Vitamin C, Vitamin E, Selenium, and Carotenoids. Washington, DC: National
Academy Press2000; 529.
92. Health Canada. Reference values for vitamins. https://www.canada.ca/en/health-
canada/services/food-nutrition/healthy-eating/dietary-reference-intakes/tables/
reference-values-vitamins-dietary-reference-intakes-tables-2005.html. Modified
November 29, 2010. Accessed May 28, 2017.
93. Bhalla A, JajooUN, Jain AP, Kalantri SP. Haemolysis with anti-malarialdrugs in glu-
cose 6 phosphate dehydrogenase deficiency. J Ayub Med Coll Abbottabad.
2004;16:75–78.
94. US Food and DrugAdministration. Food additivespermitted for direct addition to
food for human consumption. Quinine. Fed Regist. 2016. Title 21, vol. 3. Codified
at 21 CFR §172.575. Revised as of April 1, 2018.
95. Bichali S, Brault D, Masserot C, et al. Maternal consumption of quinine-containing
sodas may induce G6PD crises in breastfed children. Eur J Pediatr.
2017;176:1415–1418.
96. Lalloo DG, Shingadia D, Bell DJ, et al. UK malaria treatment guidelines 2016. J
Infect. 2 016;72:635–649 .
97. Zuccotti GV, Redaelli F, Gualdi V, et al. Hemolytic crisis in a G6PD-deficient infant after
ingestion of pumpkin. Ital J Pediatr. 2014;40:71. doi: 10.1186/1824-7288-40-71
98. Globerman H, Navok T, Chevion M. Haemolysis in a G6PD-deficient child induced
by eating unripe peaches. Scand J Haematol. 1984;33:337–341.
99. Ko CH, Li K, Ng PC, et al. Pro-oxidative effects of tea and polyphenols,
epigallocatechin-3-gallate and epigallocatechin, on G6PD-deficient erythrocytes
in vitro. Int J Mol Med.2006;18:987–994.
100. Lai SW, ChenJH, Kao WY. Acute hemolytic anemia in glucose-6-phosphate dehy-
drogenase deficiency complicated by Ginkgo biloba.ActaHaematol.
2013;130:288–290.
Nutrition Reviews
V
R
Vol. 0(0):1–11 11
Downloaded from https://academic.oup.com/nutritionreviews/advance-article-abstract/doi/10.1093/nutrit/nuy053/5151226 by Health Canada /PHAC user on 31 October 2018
