Thiamine deficiency due to sulphur dioxide preservative in 'pet meat' - A case of déjà vu

Full text

Thiamine deficiency due to sulphur
dioxide preservative in ‘pet meat’
– a case of déjà vu
R MALIKaand D SIBRAAb
This month’s Clinical Section features an excellent paper1concerning thiamine defi-
ciency in canine patients treated by Rita Singh and colleagues at the North Ryde
Veterinary Specialist Centre, Sydney. The series of cases is just what would be
expected from a state-of-the-art referral hospital in the first decade of the new millennium:
a team approach to investigation and treatment, a timely diagnosis in most patients, utili-
sation of advanced diagnostics, and outcomes that were generally favourable, even in dogs
that presented late with advanced signs.
A range of cases is described, serving to emphasise the spectrum of physical findings that
may develop in dogs with this disease. Interestingly, one gets the sense that the authors’
diagnostic acumen was refined by their experience with the index case, as a subtle pattern
of suggestive clinical findings emerged, facilitating the ‘pattern recognition’ of subsequent
cases. Magnetic resonance imaging (MRI) findings in dogs with thiamine deficiency have
been reported on only one occasion previously and the additional data provided from the
case so examined suggests that the changes are characteristic, as might be expected from
the neuroanatomical distribution of lesions (periventricular brainstem grey matter) in
carnivores with this vitamin deficiency (Figure 1). Use of the thiamine pyrophosphate
effect test in this case provided unequivocal evidence that thiamine deficiency was the
cause of the observed MRI lesions.
In cats, a diagnosis of thiamine deficiency is facilitated by the presence of more character-
istic clinical signs, especially bilateral pupillary dilatation and the so-called ‘praying sign’,
in which there is an active spastic ventroflexion of the head and neck. This sign can be
precipitated by performing a manoeuvre that accentuates the vestibular dysfunction,
namely suspending the cat by its hindquarters and ‘nose-diving’ the patient towards the
floor or tabletop. This is not to be confused with passive ventroflexion of the head and
neck, which is seen in association with diseases that cause muscle weakness, such as
hypokalaemia and myasthenia gravis.
In dogs, signs of thiamine deficiency tend to be more cryptic. It is important, therefore, to
have a high index of suspicion for this diagnosis in canine patients with neural deficits,
and sometimes even in dogs presenting with non-specific findings such as depression,
vomiting and anorexia. Neurological findings in dogs with thiamine deficiency may
include ataxia, vestibular signs, paresis, altered mental state, seizures and combinations of
the above, consistent with a multifocal disease process.
The importance of taking a careful dietary history cannot be overemphasised, as typically
this provides the vital clue that thiamine deficiency is the correct diagnosis. In Australia,
the key historical finding is feeding ‘pet mince’, ‘pet meat’ or ‘unrefrigerated food rolls’
especially when there is a relationship in time between the development of signs and a
change in the composition of the diet and/or its supplier. In the distant past, and in other
countries, feeding of fresh or canned fish containing thiaminases or inactivation of
thiamine by cooking or processing have been implicated in the aetiology, however nowa-
days commercial canned and dry food produced for cats and dogs is generally supple-
mented with thiamine. There is a suggestion from Singh’s paper, and the historical litera-
ture cited, that signs might develop more quickly in young patients, perhaps reflecting
increased requirements for thiamine during growth.
Clinical
Clinical
ClinicalClinical
EDITOR
MAUREEN REVINGTON
ADVISORY COMMITTEE
ANAESTHESIA
LEN CULLEN
AVIAN MEDICINE AND SURGERY
GARRY CROSS
EQUINE MEDICINE AND SURGERY
JOHN BOLTON
LABORATORY ANIMAL MEDICINE
MALCOLM FRANCE
OPHTHALMOLOGY
JEFF SMITH
PATHOLOGY
CLIVE HUXTABLE
PHARMACOLOGY / THERAPEUTICS
STEPHEN PAGE
PRODUCTION ANIMAL MEDICINE
JAKOB MALMO
RADIOLOGY
ROBERT NICOLL
REPRODUCTION
PHILIP THOMAS
SMALL ANIMAL MEDICINE
BRIAN FARROW
DAVID WATSON
SMALL ANIMAL SURGERY
GEOFF ROBINS
GERALDINE HUNT
WILDLIFE / EXOTIC ANIMALS
LARRY VOGELNEST
EDITORI AL ASSISTA NT AND
DESKTOP PUBLISHIN G
ANNA GALLO
CONTRI BUTIONS INVIT ED
Pra ctis ing vet erin aria ns a nd oth ers a re
invit ed to contri bute clinica l ar ticle s to th e
Aus tral ian Vet erin ary Jou rnal . We wil l
con side r m ateria l in a vari ety of form ats,
inc ludi ng cl inic ally orie ntated revi ews,
rep orts of ca se ser ies, in divi dual ca se
stu dies , d iagn osti c e xercis es, and letter s
con tain ing com ment s or queri es.
Contri butors should con sult in structi ons fo r
auth ors on
htt p: //w ww. ava .co m.a u/a vj int ro
and recent issues of t he j ournal for g uide-
lines as to formatting.
Ove r re fere ncin g sh ould b e av oided:
author s shou ld pre ferably quote only those
art icle s they fe el a re most likel y to be o f
interest and benefit to readers.
Send all contributions to:
Editor , AVJ Clinical Section
AVA House, 272 B runswick Road,
Brunsw ick, V ic 305 6
Phone: (03) 9387 298 2
Fax: (03) 9388 0112
Email: deskt op@ava. com.au
Authors, please note that from
July 1st the manuscript handling f
ee
will be $55, including GST.
Austr alian
VETERINARY
JOURNAL
C L I N I C A L S E C T I O N
408 Australian Veterinary Journal Volume 83, No 7, July 2005
aPost Graduate Foundation in Veterinary Science, The University of Sydney, New South Wales, 2006
bSolicitor and Food Safety Consu ltant, 123 Cons titution Road, West Ryde , New South Wales 2114
(Former Chief Food Inspector of New South Wales)
LEADING ARTICLE

Unfortunately, we were left with extreme frustration after reading
this paper! Why, one might ask, when it is such a ‘good paper’ -
well written, presented to an appropriate audience and with
acceptable treatment outcomes? The reason is simple: each of
these cases was preventable, including the pups that died and the
index case that afforded the opportunity to characterise the MRI
findings, albeit at considerable expense for the owners, and not
without risk to the dog because of the dangers inherent in anaes-
thetising a patient with likely increased intracranial pressure.
Moreover, thiamine deficiency was meticulously characterised in
Australia many years ago, with the cause in most patients being
determined beyond doubt. Perhaps the best Australian studies
were conducted by Professor Virginia Studdert and Dr Robert
LaBuc in Melbourne,2and Dr Robert Steele in Sydney,3who
demonstrated clearly that thiamine deficiency occurs when cats or
dogs are fed exclusively a diet containing high concentrations of
sulphur dioxide. Steele’s case was especially poignant, as the cat
developed fatal iatrogenic thiamine deficiency while being
prescribed an elimination diet of kangaroo meat to exclude the
possibility of a food-associated dermatosis. The label on the meat
fed to the cat indicated it was vacuum-packed, but did not
disclose that it contained preservatives. Steele demonstrated
further that 54 of 63 meats sold for pet consumption in 1996
contained sulphite preservatives.
Preservatives that liberate sulphur dioxide (220 - sulphur dioxide,
221 - sodium sulphite, 222 - sodium bisulphite, 223 - sodium
metabisulphite, 224 - potassium metabisulphite, 225 - potassium
sulphite, 228 - potassium bisulphite) are commonly added, in
varying degrees, to ‘pet meat/mince’ to diminish the odour
produced by bacteria that multiply in food, and delay the reduc-
tion of myoglobin, which results in the meat appearing brown
rather than red. Sulphur dioxide rapidly inactivates thiamine
present normally in meat and meat by-products, and indeed,
there may be sufficient preservative to inactivate thiamine present
in other dietary components fed concurrently, for example,
brewers yeast. It is possible to determine the presence of sulphur
dioxide in food inexpensively by adding 10 drops of a test solu-
tion (0.02% malachite green and 0.02% sodium benzoate) to a
test diet; absence of colour after 2 minutes indicates the presence
of sulphur dioxide. This test is very sensitive for even small
amounts of preservative. In addition to their effects on thiamine,
sulphites have been associated with the full range of food intoler-
ance symptoms in people, including headaches, irritable bowel
symptoms, behavioural disturbances and skin rashes. They are
also well known for their ability to exacerbate asthma in human
patients,4,5 which might be a pertinent consideration when
managing cats with ‘asthma’, or dogs with chronic bronchitis or
atopic dermatitis. In relation to this point, it should be noted that
sulphites are permitted in very large concentrations (up to 3000
mg/kg) in some foods destined for human (and therefore possibly
animal) consumption, for example dried fruits and vegetables.
A trip to a local supermarket or any large pet store or warehouse
will support the contention that there are large numbers of
suppliers of ‘pet meat’ and ‘food rolls’, and that these products
seem popular with the public. It would be interesting to know
what proportion of the pet food market is catered to by this type
of food, and whether such foods are fed exclusively, or as a part of
a heterogeneous diet. Previous studies2,3 have shown that this type
of diet may have sufficient sulphur dioxide content to destroy
endogenous thiamine present in the ration. A level of 400 mg
sulphur dioxide/kg depletes thiamine by 55%, while 1000mg/kg
depletes it by 95%.2,3 Thiamine given as a supplement concur-
rently is likewise inactivated.2,3 Considering that these foods may
be fed exclusively to cats and dogs it is indeed surprising that
thiamine deficiency does not occur more commonly, and the
reason for this is worthy of further investigation.
Clinical
Clinical
ClinicalClinical
409
Australian Veterinary Journal Volume 83, No 7, July 2005
Table 1. Sulphur dioxide content (in mg/kg) in different food items purchased on 12th April 2005 at a suburban supermarket and a pet food
wholesaler. The label claims in relation to preservatives are tabulated also.
Manufacturer Intended target species Label description of product Preservative claim on label Sulphur dioxide content
(mg/kg)a
Brand A Dog (puppies) Roll; 900 g -33
Brand A Dog Roll (cooked); 2 kg No preservatives 18
Brand B Dog Roll (beef and lamb with liver, -29
heart, cereal, vegetables); 1.5 kg
Brand B Cat Roll (chicken, lamb, cereal); 365 g - 119
Brand B Dog and Cat Bulk pet mince with kangaroo, - 640b
beef, lamb, trimming; 2.5 kg
Brand B Cat Kangaroo, lambs fry, heart; 1 kg Preservative 220 (minimal) 738b
added to retain freshness
Brand C Cat Pilchard (chunky, canned); - < 10
product of Thailand
Brand D Cat Kangaroo and brawn; 1kg - 453b
Brand D Dog and Cat Minced kangaroo; 1 kg - 356b
Brand D Dog and Cat Diced lamb - 1,058b
Brand E Cat Seafood platter (four varieties of - 45
canned seafood); Product of Thailand
Brand E Cat Mince (lean kangaroo); 1 kg Preservative (220) (minimal) 714b
Brand F Dog Roll (cooked beef, chicken, Fully cooked 14
vegetables, pasta); 2 kg
amg/kg is equivalent to parts per million (ppm)
bThese diets have the propensity to cause thiamine deficiency if fed exclusively.
- = Information not apparent on package label

destined for pet consumption - this legislation is mostly
concerned with food producing species and horses. From our
research, there appear to be no standard or legislative require-
ments specifically in relation to the contents of pet food, apart
from the Trade Practices Act which protects against manufacturers
making misleading or untrue claims. The industry is currently self
regulated via the Pet Food Industry Association of Australia. The
web page (http://www.nswfitc.com.au/a/2680.html) of this
organisation states they set industry standards via a code of prac-
tice which apparently incorporates standards for labelling,
marketing, nutrition design and claims. Details of this code of
practice are not available from the website. Pet food labelling in
Australia is governed by both state industry laws and consumer
product laws, and in some states there is a legal requirement that
the words ‘PET FOOD ONLY’ appear on the label, and some
state legislation makes it mandatory for a picture of the animal
species for which the food is intended to feature on the label. A
minimum guaranteed analysis declaration is also required stating
minimum percentage of crude protein, fat, fibre, moisture, salt
and, optionally, other ingredients. There is, however, no legislative
control over what is incorporated into pet food, and in particular,
whether preservatives are added.
In contrast, the NSW Food Authority carefully regulates butcher
shops, delicatessens, supermarkets and other food outlets to ensure
that preservatives are not added to meat products for h
uman
consumption (except sausages, sausage meat and cooked manu-
factured meats). Fines of up to $55,000 for individuals and
$275,000 for corporations apply. In view of the severe adverse
consequences that preservatives can have on companion animals,
veterinarians have an obligation to lobby that appropriate legislation
likewise be drafted to cover food intended for consumption by pet
s.
The success of the ban on tail docking has demonstrated what this
profession is capable of achieving when it presents a coherent
unified front. A similar effort is required in relation to the issue of
preservatives in pet food. Clearly, we have a responsibility to speak
out on such issues in our role as advocates for companion
animals, and to provide a scientific basis that underpins our posi-
tion. For legal and political reasons, this could best be tackled at
the state level by appropriate arms of the veterinary profession. In
situations where pressure can be brought to bear, it is likely to be
most appropriate that those organisations representing a wide
body of veterinarians, such as the Australian Veterinary
Association and Australian College of Veterinary Scientists, be
involved, so that the full force of the profession can be harnessed.
It could be effectively argued that sulphite preservatives are actu-
ally ‘drugs’, and that their inclusion in pet mince could be
pursued by the Australian Pesticides and Veterinary Medicines
Authority. Certainly, the implications of thiamine deficiency are
sufficiently severe that they should be considered a life-threat-
ening adverse drug reaction, and in the interim, this may be a
fruitful line to pursue.
What advice can we give our clients in the meantime concerning
the feeding of raw meat and meat by-products for consumption
by cats and dogs? We should advise consumers not to be seduced
by the claims on the packaging. Most owners of companion
animals wish to do the best they can for their pets. They are told
that the contents of ‘pet mince’ and ‘food rolls’ are nutritious and
provide the best quality meat for the animals. They are often
cheaper than fresh meat fit for human consumption, more conve-
nient to buy and sometimes do not require refrigeration. Owners
should be informed that because there is no legal requirement to
To provide a current estimate of the prevalence of sulphur dioxide
in pet meat, pet mince and food rolls, one of the authors (RM)
obtained a representative selection of these products from one
suburban supermarket and one regional pet food warehouse on
the 12th April 2005 and submitted them to a commercial labora-
toryafor testing. Specimens were tested by Mr Roger Mooney
using AOAC Method 962.16 (modified Monier Williams
method).6A small number of cans of commercial cat food
produced overseas were tested also. The results of these analyses,
presented in Table 1, show a high content of sulphur dioxide in
approximately half of the food items tested.
Due to the high cost of testing, only a limited number specimens
were examined. Even so, certain trends emerged. With one minor
exception, refrigerated food rolls (that had been cooked during
processing) had low concentrations of sulphur dioxide unlikely to
lead to thiamine deficiency. In contrast, all products containing
diced or minced meat had dangerous concentrations of sulphur
dioxide. Two products that stated preservative 220 was present in
‘minimal’ quantities had concentrations of sulphur dioxide in
excess of 700 mg/kg. Four products that did not advertise the
presence of preservatives had dangerous concentrations of sulphur
dioxide (ranging from 356 to 1,058 mg/kg). Canned fish foods
produced in Thailand had low concentrations of sulphur dioxide,
although we did not assay their thiamine content. We were unable
to find any unrefrigerated food rolls for testing, which was inter-
esting, as these were said historically to contain especially high
concentrations of sulphites.
The principal underlying problem is the lack of regulation over
the range of meat products and by-products sold for consumption
by pets in New South Wales. No requirement currently exists
under the NSW Stock Foods Act or the Stock Medicines Act to
identify the concentration of sulphite preservatives in meat
Clinical
Clinical
ClinicalClinical
410 Australian Veterinary Journal Volume 83, No 7, July 2005
aSilliker Microtech Pty Ltd, Unit 2C, Regents Park Estate, 391 Park Road,
Regents Park, New South Wales 2143
Figure 1. Photograph of a transversely sectioned, formalin-fixed,
brain from a cat that died of thiamine deficiency. Note the multiple,
bilaterally symmetrical, focal areas of grey-red discolouration
throughout the brainstem. Histologically, such discoloured areas
correspond to regions of haemorrhagic malacia. The extent and
severity of such lesions would account for the poor prognosis of
advanced cases of thiamine deficiency. Kodachrome kindly
supplied by Professor BRH Farrow.

state the presence of preservatives in these foods, their presence is
not invariably advertised on the label. We should explain that the
use of preservatives does not prevent spoilage or putrefaction, but
that rather it only masks the most revealing signs, namely the
unsavoury odour and brown discolouration of the meat. Finally,
we should draw pet owners’ attention to the fact that ‘pet meat’
has not necessarily been subjected to the stringent meat inspec-
tion and processing assessment that occurs for meat designated
for human consumption.
Perhaps the simplest way to ensure that thiamine deficiency does
not occur is to recommend feeding a commercially prepared,
nutritionally complete dog food, either canned and/or kibble in
type. For those clients and veterinarians convinced of the health
benefits of feeding a more natural diet, either exclusively or as a
component of a varied diet, it should be recommended that such
food be obtained from the local butcher, where appropriate legis-
lation affords protection against the use of preservatives that could
give rise to thiamine deficiency.
Acknowledgments
The authors wish to thank Sue Schreiner, who helped with the
drafting of this editorial, for her perspective as a dog owner and
research on the internet; and Paul Canfield and Erin Bell.
References
1. Singh R, Thompson M, Sullivan N, Child G. Thiamine deficiency in dogs due to
feeding of sulphite preserved meat.
Aust Vet J
2005; 83:412-417.
2. Studdert VP, Labuc RH. Thiamine deficiency in cats and dogs associated with
feeding meat preserved with sulphur dioxide.
Aust Vet J
1991;68:54-57.
3. Steel RJ. Thiamine deficiency in a cat associated with the preservation of ‘pet
meat’ with sulphur dioxide.
Aust Vet J
1997; 75:719-721.
4. Steinman HA, Weinberg EG. The effects of soft-drink preservatives on asth-
matic children.
S Afr Med J.
1986;70:404-406.
5. Picado C. Classification of severe asthma exacerbatio ns: a pro posal.
Eur
Respir J
1996;9:1775-1778.
6. Hillery BR, Elkins ER, Warner CR et al. Optimized Monier-Williams method for
determination of sulfites in foods: collaborative study.
J Assoc Off Anal Chem.
1989;72:470-475.
Clinical
Clinical
ClinicalClinical
411
Australian Veterinary Journal Volume 83, No 7, July 2005
Domestic Animal Behavior for Veterinarians and Animal Scientists. 4th edn.Houpt K, Blackwell Publishing, Iowa, 2005, 506 pages. Price $A137.
ISBN 0 8138 0334 9.
One of the guiding principles of my veterinary education was that it was necessary to understand what was normal before attempting to come to
grips with the abnormal. However, in books on behavioural problems in animals these fundamentals are often bypassed or minimised in an
eagerness to get to the ‘clinically relevant’ part of the book – descriptions of behavioural problems and their associated therapies. Thankfully, this is
not the case with Dr Houpt’s book and this remains one of its greatest strengths. Ever since the first edition in 1982, it has emphasised underlying
behavioural mechanisms and documenting normal behaviours, thus allowing behavioural problems to be placed in an appropriate context when
discussed. To the busy clinician who simply wants to know how to deal with a particular problem, such fundamentals may seem irrelevant, however,
by understanding the mechanisms that drive behavioural expression, this provides the therapist with first principles to fall back on when the common
‘cookbook’ approach to treating behavioural problems fails.
As with previous editions, the book is divided into nine chapters – communication, aggression and social structure, biological rhythms and sleep,
sexual behaviour, maternal behaviour, development of behaviour, learning, ingestive behaviour, and miscellaneous behavioural disorders. Each
chapter is organised into general principles and a comprehensive summary of what is known for each of seven major domesticated species - horses,
dogs, cats, pigs, cattle, sheep and goats (although the information on goats is rather limited). While chapter nine discusses miscellaneous problems,
the majority of behavioural disorders are discussed within the appropriate chapters under the heading of the species concerned. This organisation
makes the book a pleasure to read, the information easy to find, and the relevance of each behavioural system to specific behavioural disorders
obvious and informative.
For those readers familiar with previous editions, there are few surprises, despite the text having been comprehensively updated with contemporary
references and new information (over 1800 supporting references of which ~20% are new since the 1998 edition). The reason for this is that much of
the core work in applied animal behaviour that is central to the discussions in this book were undertaken 10 to 20 years ago.With only one exception,
all the figures can be found in previous editions of the book, thus, while the new additions improve the quality of the information, the need for
someone to update from their 2nd or 3rd edition will be more a matter of personal taste than necessity.
So who should buy this book? Anyone working with large animals will find it an excellent resource for its comprehensive summaries of large animal
behaviours and associated behavioural problems, information that is often not readily available in other texts. The companion animal practitioner who
wants a broader understanding of the behavioural processes they are attempting to modify will also gain much from this edition. I’d also recommend
it to people who enjoy being able to answer the obscure questions posed by clients, for example how many times a day should a horse urinate; how
long does a sheep sleep and what propor tion is REM (but exactly what sheep count when they go to sleep is not discussed); can animals see in
colour; the clinical relevance of the curl in a pig’s tail or the names and meanings of the 14 different vocalisations a cat makes.
Despite my high praise for the book, I do have a minor criticism. For the next edition I’d recommend that the publisher actually bother reading it (at
least to the bottom of the first page). The blurb on the back cover promises: “New to this edition are discussions of progress and research in the rela-
tionship between behavior and animal welfare”. Unfor tunately, the author states very clearly in her preface to the new edition: “Welfare is an important
topic that has been covered in several books and many publications, so it is not directly covered here.” And indeed it is not. While I can understand the
necessity of deferring to more specialised texts, as many of the issues are outside the scope of this book, a small reference to welfare might not have
gone astray, in particular, during discussions of vices and stereotypic behaviours that are thought by many to result from poor environmental condi-
tions and are signs of reduced animal welfare.The reason for mentioning this is that many traditional therapies for these problems are based on phys-
ical restraint or redirection. It would be wise to warn in these cases that simply addressing the sign, rather than the cause, may ‘fix’ the behavioural
disorder, but not improve the welfare of the animal in conditions where it is compromised.
MR Low
Matt Low is a member of the animal behaviour chapter of the Australian College of Veterinary Scientists and has recently completed his PhD in behavioural ecology at
Massey University.
BOOK REVIEW