Beynen AC, 2019. Arsenic in petfood

Abstract

Arsenic in petfood Online commentaries on detection of arsenic in pet and baby foods (cf. 1-6) probably stirred up concern among some pet parents, especially when they know arsenic as poison used in older detective stories. Arsenic is not intentionally added to petfood, but the pervasive earth element is intrinsic to many ingredients. Possibly, arsenic is an essential nutrient for dogs and cats, albeit in tiny amounts. Arsenic, often referred to as heavy metal, is toxic only in high quantities, but with its chemical form as a co-definer. European legislation has set upper limits for arsenic in complete food for dogs and cats (7). The maximum is 2.3 mg total arsenic per kg canned or kibbled food when completely dried. However, the maximum is 11.4 mg when the food contains (derivatives of) fish and/or seaweed. That higher maximum unifies practical and health considerations (8): the two aquatic products are relatively rich in arsenic, but in a form of low toxicity. Chemical analysis of total arsenic in complete petfoods indicates typical compliance with European legislation. Absorbed arsenic compounds are converted by the dog's liver and/or kidneys and then excreted with urine. Research data show that long-term intake of up to 20 mg total arsenic per kg dry weight of food is not toxic for dogs. Thus, commercial dog foods are generally safe with regard to their levels of total arsenic. In cats, the maximum tolerated dose of dietary arsenic is unknown. Accidental ingestion of products containing a toxic amount of arsenic can cause acute poisoning in dogs and cats. Such products used to be herbicides, insecticides and wood preservatives (9, 10), but today ant and roach baits are major culprits (11, 12). Symptoms are vomiting, salivation, abdominal pain, staggering gait and diarrhea. Affected pets should be taken to a veterinary facility immediately. Dietary arsenic Concentrations of total arsenic (As) in dry and wet petfoods are mostly lower than 2.3 mg/kg dietary dry matter (ddm) (13-19, Notes 1 and 2). Higher values are only found in foods containing fish derivatives (8, 13, 18, 20, Note 3). Percent inclusion of fish and As in food were positively related (20). Total As in fish-based dog and cat foods normally is well below 11.4 mg/kg ddm (8, 13, 18, 20), but a higher level is sometimes found (8, 20). Thus, the European allowable limits are usually not exceeded. As is ubiquitous and thus may be found at detectable amounts in petfood ingredients. Moreover, marine fish, seaweed and rice concentrate As from the environment. Fish meal and oil, hijiki seaweed and rice may have total As contents of up to 20 and 14 (21), 34 (22) and 2 (23) mg/kg dry weight. In finfish, As mainly occurs as arsenobetaine, whereas fatty fish and seaweed contain arsenolipids (21, 24) and arsenosugars (22). Rice contains inorganic As, but speciation is variable (25).

Full text

Creature Companion 2019; April: 44-45.
Anton C. Beynen
Arsenic in petfood
Online commentaries on detection of arsenic in pet and baby foods (cf. 1-6) probably stirred up
concern among some pet parents, especially when they know arsenic as poison used in older
detective stories. Arsenic is not intentionally added to petfood, but the pervasive earth element is
intrinsic to many ingredients. Possibly, arsenic is an essential nutrient for dogs and cats, albeit in
tiny amounts. Arsenic, often referred to as heavy metal, is toxic only in high quantities, but with its
chemical form as a co-definer.
European legislation has set upper limits for arsenic in complete food for dogs and cats (7). The
maximum is 2.3 mg total arsenic per kg canned or kibbled food when completely dried. However,
the maximum is 11.4 mg when the food contains (derivatives of) fish and/or seaweed. That higher
maximum unifies practical and health considerations (8): the two aquatic products are relatively
rich in arsenic, but in a form of low toxicity.
Chemical analysis of total arsenic in complete petfoods indicates typical compliance with European
legislation. Absorbed arsenic compounds are converted by the dog’s liver and/or kidneys and then
excreted with urine. Research data show that long-term intake of up to 20 mg total arsenic per kg
dry weight of food is not toxic for dogs. Thus, commercial dog foods are generally safe with regard
to their levels of total arsenic. In cats, the maximum tolerated dose of dietary arsenic is unknown.
Accidental ingestion of products containing a toxic amount of arsenic can cause acute poisoning in
dogs and cats. Such products used to be herbicides, insecticides and wood preservatives (9, 10), but
today ant and roach baits are major culprits (11, 12). Symptoms are vomiting, salivation,
abdominal pain, staggering gait and diarrhea. Affected pets should be taken to a veterinary
facility immediately.
Dietary arsenic
Concentrations of total arsenic (As) in dry and wet petfoods are mostly lower than 2.3 mg/kg dietary
dry matter (ddm) (13-19, Notes 1 and 2). Higher values are only found in foods containing fish
derivatives (8, 13, 18, 20, Note 3). Percent inclusion of fish and As in food were positively related
(20). Total As in fish-based dog and cat foods normally is well below 11.4 mg/kg ddm (8, 13, 18, 20),
but a higher level is sometimes found (8, 20). Thus, the European allowable limits are usually not
exceeded.
As is ubiquitous and thus may be found at detectable amounts in petfood ingredients. Moreover,
marine fish, seaweed and rice concentrate As from the environment. Fish meal and oil, hijiki
seaweed and rice may have total As contents of up to 20 and 14 (21), 34 (22) and 2 (23) mg/kg dry
weight. In finfish, As mainly occurs as arsenobetaine, whereas fatty fish and seaweed contain
arsenolipids (21, 24) and arsenosugars (22). Rice contains inorganic As, but speciation is variable
(25).

Metabolism
Three hours after a single oral dose of 74As as arsenic acid (H3AsO4), about 90% of the radioactivity
was present in dog’s blood (26), pointing to high absorption. In contrast, dogs poorly absorbed As
from arsenate bound to bog ore (27). Dogs excrete negligible amounts of As into bile (28). However,
the canine kidney reduces arsenate to arsenite for excretion (29, 30) and also excretes As as
dimethylarsinic acid (DMA) (26, 30-32).
As ingested in the form of arsenate or arsenic sulfide is converted into DMA by dogs (26, 30-32).
Biomethylation of inorganic As, which takes place in the dog’s liver (33), appears unreported for
cats. Interestingly, arsenic oxide was highly toxic for a feline kidney cell line, but DMA was not (34).
Toxicity
The results of five toxicity studies (35-39, Note 4), indicate that long-term intake of 20 mg total As/kg
ddm is not toxic in dogs. Inorganic As sources were tested, except for one study (36). The highest,
analysed value reported for commercial petfood was 18.9 mg/kg ddm, which concerned a fish-
based, wet food (20). Two-year toxicity studies in dogs (36, 37, 39) did not mention cancer
development. Toxicity feeding studies in cats are lacking. Intravenous administration of arsenic
trioxide, as uncorrected equivalent of 26 mg As/kg ddm for 6 days, was toxic in cats (40).
Toxicity of trivalent arsenicals, such as arsenite, relates to their reaction with thiol-containing
molecules. In dogs, administration of lipoic acid, a dithiol and co-factor of various enzymes,
prevented arsenite toxicity (41). Harm by pentavalent arsenate may result from replacing phosphate
in biochemical reactions. In rats, dietary inorganic phosphate partly counteracted arsenate toxicity
(42).
Oral toxicity of arsenicals in descending order, on the basis of acute LD50 values in rodents, is as
follows: arsenite, arsenate, DMA, arsenobetaine (43). Arsenobetaine is non-toxic due to readily
excretion in the urine intact. Arsenolipids and arsenosugars may be metabolized into di-methylated
arsenicals that could exert toxic effects (44, 45). Canine and feline metabolism and toxicity of those
dietary, organic arsenic forms are unknown.
Arsenic exposure
Dogs that ingested sodium arsenite for four months, at three intake levels including zero, showed
dose-dependent As concentrations in urine, liver and hair (46). Urine may be an indicator of recent
dietary As exposure, whereas hair reflects the longer term. Hair As was very high in dogs from an
area with As-rich groundwater (47, 48). Liver As did not differ between dogs from rural and urban
areas (49).
Note 1
Total arsenic in complete dog foods (mg/kg dietary dry matter)
Ref
Year
Ana
-
lysis
Wet
dog
foods
Dry
dog foods
n
Mean
Range
n
Mean
Range
13
1976
a
27$
0.22
<0.20
-
0.67
14
2011
b*
13$
0.20
0.10
-
0.45
18$+
0.15
0.03
-
0.28

15
2012
c
34
0.39
^
<0.29
-
0.49
16
2013
d
18$+
0.19
0.06
-
0.79
17
2017
nr
297$
0.15
nr
18
2018
d=
51
0.58#
0.03
-
4.40
19
2018
d
6
0.85
0.55
-
1.50
62
0.19
0.08
-
1.00
Overall mean/range
0.42
0.10
-
1.50
0.28
0.03
-
4.40
Total arsenic in complete cat foods (mg/kg dietary dry matter)
Ref
Year
Ana
-
lysis
Wet cat
foods
Dry
cat foods
n
Mean
Range
n
Mean
Range
13
1976
a
4$
1.68
0.90
-
2.30
14
2011
b*
14$
0.26
0.06
-
0.48
13$
0.17
0.08
-
0.31
17
2017
nr
121$
0.09
nr
19
2018
d
6
1.95
0.80
-
4.55
14
0.20
0.10
-
0.33
Overall mean/range
1.30
0.06
-
4.55
0.15
0.08
-
0.33
Ref, Year = reference and year of publication; nr = not reported; *Dry and wet foods were assumed
to contain 90 and 20% dry matter; = results are expressed as mg arsenic/Mcal metabolizable energy;
it is assumed here that 1 Mcal equals 0.251 kg dietary dry matter; $ presumably complete foods; +
identical samples; ^middle of range; # mean of medians for three food groups each comprising 17
products.
Analysis: a, samples were subjected to dry ashing, distillation of arsine and determination using the
silver diethyldithiocarbamate spectrophotometric method; b, food samples were ground in a
cryogenic freezer mill (dry samples) or blender (wet samples), digested with concentrated nitric acid,
and analyzed for arsenic by inductively coupled plasma-mass spectrometry (ICP-MS); c, instrumental
neutron activation analysis; d, microwave-assisted nitric acid digestion and ICP-MS
Note 2
Other studies than those given in the table above also report arsenic levels in petfood, but they do
so without classifying the values according to nutritional completeness, moisture content and target
species. The Clean Label Project (50) reports arsenic values for 1,084 pet products from 80 brands,
but only disclosed the average values for dry dog and cat foods (17) as presented by the table above.
In addition, the average amounts for a certain brand were presented, which amounted to 1.18 (n= 3)
and 0.93 (n= 10) mg/kg ddm for dry cat and dog foods (17). A manufacturer reported the average
arsenic content of its dry foods, which was found to be 0.43 mg/kg ddm; further details are not
provided (51). The Alltech Heavy Metal Survey (52) includes petfood that is not specified while the
analysed arsenic concentrations are not given separately.
Squadrone et al. (8) have analysed arsenic in food samples during the period of 2007-2012. Arsenic
was quantified by using a wet digestion method and electro thermal atomic absorption. There were
133 samples of complete and 67 samples of complementary dog and cat foods, presumably dry
foods. The totals of fish-and meat-based products were 137 and 63. Only foods containing fish or
fish derivatives had detectable amounts of total arsenic. The mean, lowest and highest arsenic
concentrations were 2.05, 0.86 and 12.50 mg/kg food.

Note 3
As mentioned, the paper by Squadrone et al. (8) clearly shows that fish-based foods are associated
with higher contents of total arsenic. The same observation was reported by Furr et al. (13) in 1976.
Two, more recent studies (18, 20) also indicate that fish is a major arsenic source.
Kim et al. (18) classified dry dog foods as fish, red meat (beef, pork, venison, bison) or poultry
(chicken, turkey, duck) based. There were 17 products from each category. The mean arsenic
contents were 0.15, 0.22 and 1.37 mg/kg ddm for the meat-, poultry- and fish-based foods. The
table above provides further information on the study.
Davies et al. (20) analysed total arsenic by ICP-MS in 113 complete cat foods (wet, n = 48; dry, n =
65) and 64 complete dog foods (wet, n = 49; dry, n = 15). The highest values were found in foods
with fish as main declared ingredient. For dry foods, the mean and highest values were 0.85 and 4.46
mg/kg ddm, while for wet foods these values were 3.15 and 18.85 mg. There was a positive
association between percentage fish and arsenic in food. Assessment of the percent of fish
derivatives in the foods is not described. Dry and wet foods without fish generally contained less
than 0.5 mg arsenic/kg ddm.
Note 4
Overview of controlled experiments on arsenic intake and toxicity in dogs
Ref
Arsenic form
Breed
n
1
Duration,
years
mg As/kg dietary dry matter
2
Non
-
toxic
Toxic
35
Arsenic trioxide, As
2
O
3
mixed
2, 3
0.5
-
1.0
0, 30, 119
36
3N
4
HPA
3
Beagles
6
2.0
0, 16
32
4
37
Sodium arsenite, NaAsO
2
Beagles
6
2.0
0, 6, 28, 56
139
5
,,
Sodium arsenate, Na
3
AsO
4
,,
,,
,,
0, 6, 28,
56
139
6
38
Sodium arsenite, NaAsO
2
Beagles
6
0.5
0
85, 170, 340
7,8
39
Arsenite, AsO
2
-
?
?
2.0
125
7,9
1n = number of dogs per treatment; 2It was assumed that dry foods contained 90% dry matter
33-nitro-4-hydroxyphenyl arsonic acid: roxarsone, which has been used as growth enhancer and
coccidiostat in poultry; 4body weight was reduced; the dogs started with 64 mg As/kg ddm, which
caused loss of condition; after four weeks, the dogs were fed the unsupplemented diet and after
another four weeks they received an As level of 32 mg/kg ddm through the remainder of the two-
year testing period; 5all six dogs lost much weight and died; 6one of six dog lost weight and died;
7assumed dry food intake was 15 g/kg body weight per day; 8until day 59 of the experiment, half of
the dosages were fed; from days 59 to 183, the indicated dosages were fed; body weight was
unaffected by arsenic, but serum aspartate aminotransferase and alanine aminotransferase were
elevated in the two highest and all dose groups, respectively; 9dogs showed evidence of
gastrointestinal injury.
Literature
1. Leigh H. Is your pet’s food toxic? (April 27, 2017). https://www.abcactionnews.com/news/is-your-
pets-food-toxic

2. Dobias P. How to avoid foods high in arsenic and keep your dog healthy (March 28, 2016).
https://peterdobias.com/blogs/blog/113943301-how-to-avoid-foods-high-in-arsenic-and-keep-your-
dog-healthy
3. Fletcher L. Long-term health effects of pet food containing heavy metals: are they safe? (May 1,
2018). https://wjla.com/features/7-on-your-side/pet-food-heavy-metals
4. Arsenic in 9 brands of infant cereal. https://www.healthybabycereals.org/
5. Hirsch J. Heavy metals in baby food: what you need to know (August 16, 2018).
https://www.consumerreports.org/food-safety/heavy-metals-in-baby-food/
6. McDowell M. Report: Popular baby food brands contain high levels of arsenic, other heavy metals
(August 31, 2018). https://www.delish.com/food-news/a22885884/baby-food-arsenic-gerber/
7. Commission Regulation (EU) No 1275/2013 of 6 December 2013 amending Annex I to Directive
2002/32/EC of the European Parliament and of the Council as regards maximum levels for arsenic,
cadmium, lead, nitrites, volatile mustard oil and harmful botanical impurities. OJEU L 328/86
(7.12.2013)
8. Squadrone S, Brizio P, Simone G, Benedetto A, Monaco G, Abete MC. Presence of arsenic in pet
food: a real hazard? Vet Italiana 2017; 53 (4). doi: 10.12834/VetIt.530.2538.2
9. Selby LA, Case AA, Osweiler GD, Hayes Jr HM. Epidemiology and toxicology of arsenic poisoning in
domestic animals. Environm Health Perspect 1977; 19: 183-189.
10. Bruere SN. Arsenical poisoning in farm dogs. N Z Vet J 1980; 10: 220.
11. Gollakner R. Arsenic poisoning in cats (2018). https://vcahospitals.com/know-your-pet/arsenic-
poisoning-in-cats
12. Arsenic poisoning in dogs.
https://www.petmd.com/dog/conditions/digestive/c_dg_arsenic_poisoning
13. Furr AK, Bache CA, Gutenmann WH, Pakkala IS, Lisk DJ. Element and chlorinated hydrocarbon
content of commercial pet foods. Cornell Vet 1976; 66: 512-526.
14. Atkins P, Ernyei L, Driscoll W, Obenauf R, Thomas R. Analysis of toxic trace metals in pet foods
using cryogenic grinding and quantitation by ICP-MS, Part I. Spectroscopy 2011; 26 (1): 46-56.
15. Elias C, De Nadai Fernandes EA, Arruda Bacchi M. Neutron activation analysis for assessing
chemical composition of dry dog foods. J Radioanal Nucl Chem 2012; 291: 245-250.
16. Kelly DG, White SD, Weir RD. Elemental composition of dog foods using nitric acid and simulated
gastric digestions. Food Chem Toxicol 2013; 55: 568-577.
17. Orijen averages (April 28, 2017) https://www.cleanlabelproject.org/blog-orijen2/

18. Kim H-T, Loftus JP, Mann S, Wakshlag JJ. Evaluation of arsenic, cadmium, lead and mercury
contamination in over-the-counter available dry dog foods with different animal ingredients (red
meat, poultry, and fish). Front Vet Sci 2018; 5: 264 doi: 10.3389/fvets.2018.00264
19. Paulelli ACC, Martins Jr AC, Silva de Paula E, Souza JMO, Carneiro MFH, Júnior FB, Batista BL. Risk
assessment of 22 chemical elements in dry and canned pet foods. J Verbrauch Lebensm 2018;
https://doi.org/10.1007/s00003-018-1178-5
20. Davies M, Alborough R, Jones L, Davis C, Williams C, Gardner DS. Mineral analysis of complete
dog and cat foods in the UK and compliance with European guidelines. Sci Rep 2017; 7: 17107 DOI:
10.1038/s41598-017-17159-7
21. Sloth JJ, Julshamn K, Lundebye A-K. Total arsenic and inorganic arsenic content in Norwegian fish
feed products. Aquaculture Nutr 2005; 11: 61-66.
22. Rose M, Lewis J, Langford N, Baxter M, Origgi S, Barber M, MacBain H, Thomas K. Arsenic in
seaweed – Forms, concentration and dietary exposure. Food Chem Toxicol 2007; 45: 1263-1267.
23. Rahman MA, Hasegawa H. High levels of inorganic arsenic in rice in areas where arsenic-
contaminated water is used for irrigation and cooking. Sci Total Environ 2011; 409: 4645-4655.
24. Sele V, Sloth JJ, Lundebye A-K, Larsen EK, Berntssen MHG, Amlund H. Arsenolipids in marine oils
and fats: A review of occurrence, chemistry and future research needs. Food Chem 2012; 618-630.
25. Jackson B, Punshon T. Recent advances in the measurement of arsenic, cadmium, and mercury in
rice and other foods. Curr Environm Health Rep 2015 March 1; 2(1): 15-24.
26. Charbonneau SM, Tam GKH, Bryce F, Zawidzka Z, Sandi E. Metabolism of orally administered
inorganic arsenic in the dog. Toxicol Lett 1979; 3: 107-113.
27. Groen K, Vaessen HAMG, Kliest JJG, De Boer JLM, Van Ooik T, Timmerman A, Vlug RF.
Bioavailability of inorganic arsenic from bog ore-containing soil in the dog. Environ Health Perspect
1994; 102: 182-184.
28. Klaassen CD. Biliary excretion of arsenic in rats, rabbits and dogs. Toxicol Appl Pharmacol 1974;
29: 447-457.
29. Ginsburg JM. Renal mechanism for excretion and transformation of arsenic in the dog. Am J
Physiol 1965; 208: 832-840.
30. Tsukamoto H, Parker HR, Peoples SA. Metabolism and renal handling of sodium arsenate in
dogs. Am J Vet Res 1983; 44: 2331-2335.
31. Tam KH, Charbonneau SM, Bryce F, Lacroix G. Separation of arsenic metabolites in dog plasma
and urine following intravenous injection of 74As. Anal Biochem 1978; 86: 505-511.
32. Zhang Y, Qiang S, Sun J, Song M, Hang T. Liquid chromatography-hydride generation-atomic
fluorescence spectrometry determination of arsenic species in dog plasma and its application to a
pharmacokinetic study after oral administration of Realgar and Niu Huang Jie Du Pan. J Chromatog B
2013; 917-918: 93-99.

33. Drobná Z, Walton FS, Harmon AW, Thomas DJ, Stýblo M. Interspecies differences in metabolism
of arsenic by cultured primary hepatocytes. Toxicol Appl Pharmacol 2010; 245: 47-56.
34. Elia AC, Magara G, Caruso C, Masoero L, Prearo M, Arsieni P, Caldaroni B, Dörr AJM, Scoparo M,
Salvati S, Brizio P, Squadrone S, Abete MC. A comparative study on subacute toxicity of arsenic
trioxide and dimethylarsinic acid on antioxidant status in Crandal Rees feline kidney (CRFK), human
heptocellular carcinoma (PLC/PRF/5), and epithelioma papulosum cyprini (EPC) cell lines. J Toxicol
Ernvironm Health, Part A 2018. DOI: 10.1080/15287394.2018.1442758
35. Calvery HO, Laug EP, Morris HJ. The chronic effects on dogs of feeding diets containing lead
acetate, lead arsenate, and arsenic trioxide in varying concentrations. J Pharmacol Exptl Therap
1938; 64: 364-387.
36. Prier RF, Nees PO, Derse PH. The toxicity of an organic arsenical, 3-nitro-4-hydroxylarsonic acid.
II. Chronic toxicity. Toxicol Appl Pharmacol 1963; 5: 526-542.
37. Byron WR, Bierbower GW, Brouwer JB, Hansen WH. Pathologic changes in rats and dogs from
two-year feeding of sodium arsenite or sodium arsenate. Toxicol Appl Pharmacol 1967; 10: 132-147.
38. Neiger RD, Osweiler GD. Effect of subacute low level dietary sodium arsenite on dogs. Fundam
Appl Toxicol 1989; 13: 439-451.
39. Scientific opinion on arsenic in food. EFSA Panel on Contaminants in the Food Chain (CONTAM).
EFSA J 2009; 7(10): 1351
40. Cheng Y, Xue J, Jiang H, Wang M, Gao L, Ma D, Zhang Z. Neuroprotective effect of resveratrol on
arsenic trioxide-induced oxidative stress in feline brain. Human Exp Toxicol 2014; 33: 737-747.
41. Grunert RR. The effect of DL-α-lipoic acid on heavy-metal intoxication in mice and dogs. Arch
Biochem Biophys 1960; 86: 190-194.
42. Majumdar S, Karmakar S, Maiti A, Choudhury M, Ghosh A, Das AS, Mitra C. Arsenic-induced
hepatic mitochondrial toxicity in rats and its amelioration by dietary phosphate. Environm Toxicol
Pharmacol 2011; 31: 107-118.
43. Hedegaard RK, Sloth JJ. Speciation of arsenic and mercury in feed: why and how? Biotechnol
Agron Sco Environm 2011; 15 (S1): 45-51.
44. Feldmann J, Krupp EM. Critical review or scientific opinion paper: Arsenosugars - a class of
benign arsenic species or justification for partly speciated arsenic fractionation in foodstuffs. Anal
Bioanal Chem 2011; 399: 1735-1741.
45. Thomas DJ, Bradham K. Role of complex organic arsenicals in food in aggregate exposure to
arsenic. J Environ Sci 2016; 49: 86-96.
46. Neiger RD, Osweiler GD. Arsenic concentrations in tissues and body fluids of dogs on chronic low-
level dietary sodium arsenite. J Vet Diagn Invest 1992; 4: 334-337.

47. Rodriguez Castro Ma C, Adreano V, Custo G, Vázquez C. Potentialities of total reflection X-ray
fluorescence spectrometry in environmental contamination: Hair of owned dogs as sentinel of
arsenic exposure. Microchem J 2013; 110: 402-406.
48. Zaccaroni A, Corteggio A, Altamura G, Silvi M, Di Vaia R, Formigaro C, Borzacchielo G. Elements
levels in dogs from “triangle of death” and different areas of Campania region (Italy). Chemosphere
2014; 108: 62-69.
49. López-Alonso M, Miranda M, García-Partida P, Cantero F, Hernández J, Benedito JL. Use of dogs
as indicators of metal exposure in rural and urban habitats in NW Spain. Sci Total Environm 2007;
372: 668-675.
50. Clean Label Project, Pet food project summary 2017: https://www.cleanlabelproject.org/pet-
food/
51. Aldolphe J. Heavy metals in pet food. https://www.petcurean.com/blog/heavy-metals-pet-food/
52. Alltech Heavy Metal Survey.
http://global.alltech.com/sites/default/files/documents/2015%20heavymetal%20survey%20A3.pdf