CHAPTER 3
SCIENTIFIC BASIS FOR THE PROTOCOL
Overview
Industry concerns
AQIS response to industry concerns
Committee's conclusions
Overview
3.1 As noted in Chapter 2, AQIS has determined time and temperature parameters
for cooking chicken meat in order to inactivate avian viruses such as
Newcastle disease (ND) and Infectious Bursal disease (IBD). AQIS based
this protocol on Dr Dennis Alexander's study of IBD. Mr Paul Hickey advised
the Committee that AQIS selected this study as a basis for development
of quarantine protocols because it represented a more conservative approach
to inactivating viruses using heat than some other studies.(1)
3.2 AQIS argued that IBD can be used as a standard for heat treatment
of other viruses because of its considerable resistance to heat treatment.
IBD is recognised as being more resistant to heat treatment than ND or
other pathogens such as salmonella bacteria.
3.3 Mr Hickey advised the Committee that there is broad agreement in
the scientific community concerning the scientific work underpinning this
protocol. Mr Hickey told the Committee that the following persons and
organisations have collectively agreed to AQIS's scientific and technical
parameters:
- veterinary officers from all of AQIS;
- Australian Chief Veterinary Officer;
- chief veterinary officers of the states;
- Australian Veterinary Association;
- veterinary officers from the Australian Animal Health Laboratory in
Geelong; and
- veterinary officers from the Bureau of Resource Sciences.(2)
3.4 Dr Sarah Kahn reinforced Mr Hickey's comments and advised the Committee
that:
...there is very broad agreement in the scientific community and even
on the part of industry that the fundamental scientific work underpinning
this quarantine protocol is agreed.(3)
Industry concerns
3.5 Without exception, all witnesses agreed that the work done by Dr
Dennis Alexander on the thermal inactivation of IBD was sound and highly
professional. However, several witnesses questioned the validity of a
number of the conclusions that AQIS had drawn on the basis of Alexander's
work.
3.6 Industry has encountered considerable difficulty in challenging AQIS's
conclusions about the adequacy of the science that underpins the cooking
protocol. Mr Tim Luckhurst explained the difficulty the industry faced:
...we have not been able to provide, if you like, scientifically based
contra evidence. We certainly raised our doubts about the assumptions
that were made, but we have not been able to provide scientific contra
evidence that says that that is incorrect.(4)
3.7 Mr Luckhurst maintained that if a procedure cannot be scientifically
proven to be unsafe, AQIS assumes that it is safe.
3.8 The areas that industry disagrees with AQIS about the science underpinning
AQIS's conclusions fall into the following categories:
- Lack of data about ND; and
- Protective effect of different heating media
3.9 The Committee also received evidence from Mr Bob Baldwin MP, Member
for Paterson, who raised concerns about AQIS's use of Dr Alexander's study.
Mr Baldwin argued that AQIS had misused Alexander's test results and tabled
a letter he had received from Dr Alexander concerning AQIS's use of his
study.(5)The letter is attached at Appendix 6.
Lack of data about ND
3.10 Dr Margaret MacKenzie, General Manager, Technical Services, Inghams
Enterprises, told the Committee that AQIS based its entire quarantine
assessment on the work of Dr Dennis Alexander and that "Alexander
himself totally disagrees with the way AQIS has used his information".
She said that AQIS had extrapolated Alexander's findings concerning the
inactivation of IBD and that "you cannot do that".(6)
3.11 Dr MacKenzie argued that much more work has to be done on the thermostability
of ND strains. She told the Committee that there are many different strains
of ND and IBD, but only a small number had been tested for thermostability:
They are innumerable. There are hundreds of different strains and many
of them have never been isolated and most of them have never been studied
for thermostability... There are probably 10 out of maybe 1,000 that
have been looked at.(7)
3.12 Dr MacKenzie pointed out that the mild strain of ND that is present
in Australia is "very heat resistant" and would survive heating
at 56C for nine hours. She emphasised that the characteristics of the
other strains of ND were unknown:
This is a mild nonpathogenic strain. What strains are there around
the world,and we have not tested them,that are going to survive the
sorts of timetemperatures that they have proposed? And we do not know
that; nobody knows that.(8)
3.13 In correspondence following the public hearing process, Dr MacKenzie
advised the Committee that virulent strains of IDB have "decimated"
the poultry industry in some parts of Thailand and have also been present
in some parts of the USA. She said that no research has been done on the
thermal inactivation of these strains.(9)
3.14 Mr Luckhurst concluded that it was misleading for AQIS to draw general
conclusions about the effectiveness of a cooking process on the basis
of an experiment conducted with one strain of IBD:
This assumes, of course, that the IBD virus that they used in that
laboratory test was, in fact, the most hardcase situation that one would
be likely to face. Secondly, of course, it assumes that the results
obtained in the laboratory would be equal to those obtained in a commercial
cooking process.(10)
Protective effect of different heating media
3.15 Industry representatives also argued that Alexander's test results
may not give an accurate indication about how ND virus would be affected
by heating in chicken meat. Dr MacKenzie explained that there is very
little information available concerning the ability of Newcastle disease
virus (NDV) to survive heating in chicken meat. She noted that for example,
if the cooking takes place in a high fat medium, the viruses and bacteria
are much more protected.
3.16 Dr MacKenzie submitted that it is not valid to draw wider conclusions
about results obtained in one test medium and that work should be done
on the characteristics of the virus when heated in chicken meat. She contended
that:
...every microbiologist knows that you cannot interpret from one medium
the results that happen on thermostability to another medium, because
there are factors in all the different media that protect differently
in temperature against cooking.(12)
3.18Dr Paul Gilchrist, a consultant for the Australian Chicken Growers
Association, supported Dr MacKenzie's comments. He told the Committee
that Dr Dennis Alexander's work is a "good standard laboratory procedure."
However, he considered AQIS's use of Alexander's work was based on two
questionable assumptions:
- any virus which was generally less susceptible to temperaturetime
effects than infectious bursal disease virus would be controlled automatically
by controlling infectious bursal disease virus; and
- the culture media in which that test was conducted would be representative
of any sort of medium in which the material was found.
3.18 Dr Gilchrist told the Committee that it was difficult to persuade
people that such assumptions were valid:
If you want to make assumptions and extrapolations, as AQIS has done,
that is fine, but people like the people in the back of this room, whose
livelihood is concerned in all of this, are not prepared to accept those
sorts of assumptions.(13)
Alexander's view of AQIS's use of his test results
3.19 In his letter to Mr Baldwin, MP, Dr Alexander advised that he had
problems with AQIS's interpretation of his data in two areas. Firstly,
he expressed concerns about the extrapolation of data to temperatures
that had not been tested. Secondly, the probability of survival of virus
after heat treatment depends on the starting infectivity level. He said
that this could confound stating specific times for particular temperatures
if not taken into account when defining acceptable treatment processes.
3.20 Alexander confirmed Dr MacKenzie's assertions that no tests had
been done on heat inactivation of ND in chicken meat and that ND may be
present in the muscles and organs of infected birds:
This problem has arisen several times in the past and I have looked
at the literature again - as far as I can tell no work has been done
on the heat inactivation of virus in chicken meat. There is good evidence
that ND virus will be present in the muscles and other organs of infected
chickens and this is particularly important in vaccinated birds, which
may have virulent virus present if infected despite failing to show
clinical signs.(14)
3.21 However, much of Dr Alexander's letter supported AQIS's views. He
confirmed that IBD is a very heat resistant virus and noted that when
conducting his experiments, he was surprised at just how heat resistant
this virus proved to be.
3.22 He advised that comparisons between heat resistance of IBD and ND
are difficult as "parallel experiments under identical conditions
are not available". However, he said that a "rough guide"
could be obtained from an experiment conducted on liquid whole egg in
his laboratory. This experiment demonstrated that ND virus lost infectivity
more rapidly and at lower temperatures than IBD when compared to the results
he had obtained with IBD.
3.23 Dr Alexander acknowledged that different testing media (egg instead
of homogenised bursae) could have affected the results obtained. However,
he considered that these would have to be "extremely different in
severity" to have produced such significant differences in heat resistance
between the two viruses. He concluded that:
...the evidence available...suggests that under comparable conditions
IBD would prove to be considerably more resistant to heat than ND.(15)
3.24 Alexander considered that laboratory test results could be extrapolated
for wider use. However, he cautioned that:
...some assessment of the closeness of the conditions in the laboratory
and the commercial processing should be made and the extent to which
variations are likely to affect the desired outcome.(16)
AQIS response to industry concerns
3.25 Mr Hickey strongly refuted suggestions that the cooking protocol
AQIS had proposed is based on inadequate science. He said that AQIS had
not relied solely on Dr Alexander's test results but has also considered
eight studies on the thermal inactivation of NDV and five studies on IBD
virus early in the risk analysis process. He asserted that AQIS had adopted
this study because it was more conservative than other studies in the
literature. Accordingly, AQIS decided to use Alexander's data as a basis
for development of quarantine protocols on the basis of that conservatism.(17)
3.26 Mr Hickey tabled AQIS's 1991 discussion paper which contained the
comparative studies conducted on the two viruses. Tables 1 and 2 below
summarise the results of these studies.
Table 1: Summary of results of thermal inactivation studies of
Newcastle disease virus
REFERENCE RESULTS
Farinas (1930) Inactivated between 50C and
55C for 30 minutes
Bushness and Erwin (1950) Inactivated between:
58C - 64C for 30 minutes
67C for 5 minutes
Foster and Thompson 4 strains, inactivated
70C for 40-50 seconds
65C for 80-90 seconds
60C for 6-7 minutes
55C for 60-80 minutes
Digioia et al (1970) 99% loss of infectivity at
43.3C
for 75 minutes
Thacore and Younger (1971) Inactivated at 55C after
10 minutes
Estola (1974) Finnish strain survived 56c
for 270 minutes
Lomniczi (1975) 24 strains, most survived 50C
for 60 minutes
Ideris et al (1990) V4 strain, survived 56C
for 9 hours
Source: AQIS discussion paper, The importation of fresh frozen
and cooked chicken meat and products from USA, Denmark,
Thailand and New Zealand, p. 12.
Table 2: Summary of results of thermal inactivation studies of
Infectious Bursal disease virus
REFERENCE RESULTS
Landgraf et al (1967) Survived 60C for 30 minutes
Inactivated 70C for 30
minutes
Cho and Edgar (1969) Survived:
60C for 90 minutes
25C for 21 days
-20C for 3 years
Fahey (1988) Survived 72C for 10 minutes
Inactivated 82C for 30
minutes
(Virus in tissue culture
fluids)
Digioia et al (1970) 99% loss of infectivity at
43.3C
for 75 minutes
MacKenzie and Spradbrow Inactivated at:
(1988) 75C for 5 minutes
80C for 1 and 5 minutes
85C for 1 and 5 minutes
(virus in meat nugget mixture)
Alexander (1988) To reduce to a probability of
0.1, that an infectious virus
remained, heat at:
70C for 50 minutes
80C for 8.8 minutes
To reduce to probability of
.001 heat at:
70C for 90 minutes
80C for 14.4 minutes
(Virus in bursa homogenate)
Source: AQIS discussion paper, The importation of fresh frozen
and cooked chicken meat and products from USA, Denmark,
Thailand and New Zealand, p. 13.
3.27 Mr Hickey assured the Committee that AQIS's proposed cooking protocol
was far more stringent than suggested as being required by any of the
research studies examined. For example, MacKenzie's 1988 study reported
IBVD as being killed in chicken nugget meat after one to five minutes
at 80C, whereas AQIS's proposal calls for processing at this temperature
for 15 minutes, that is, three times as long.
3.28 Mr Hickey attacked Dr MacKenzie's statements about the effectiveness
of the heat treatment process. He said the Dr MacKenzie used the results
of her research to support an application by Ingams to export cooked chicken
nuggets and chicken smallgoods to New Zealand and suggested that Dr MacKenzie's
concerns lacked consistency with her own research:
Yet Dr MacKenzie appears not to accept the efficacy of AQIS's proposed
protocol, although it is far more stringent than the time or temperature
parameters that apparently was proposed that New Zealand authorities
accept. In other words, 70C for five minutes for exports to New Zealand
is okay but 70C for ninety five minutes for imports is not.(18)
3.29 Mr Hickey dismissed Dr MacKenzie's concerns about possible heat
resistant strains of NDV:
We are not proposing to admit product cooked at 56C. AAHL and the BRS
have confirmed that the core temperature or cooking times proposed by
AQIS protocols would inactivate NDV if present in chicken muscle with
an extremely large margin of safety.(19)
Use of IBD as a reference organism
3.30 Mr Hickey advised the Committee that AQIS's decision to use IBD
as a reference organism was a valid approach confirmed by published data.
He noted the advice from Dr Alexander which stated that:
It is always preferable to have data from experiments run in parallel
under identical conditions, but in their absence the evidence available,
both that quoted above and published work, suggests that under comparable
conditions IBDV would prove to be considerably more resistant to heat
than NDV.(20)
Lack of data about ND
3.31 Mr Hickey advised that there have not been specific studies on the
inactivation of NDV in chicken meat. (21)However, he
considered that it is not necessary to commission such research:
While specific studies on inactivation of NDV in chicken meat have
not been conducted, AQIS considers that it is not necessary to commission
such research as there is sufficient literature to guide us. AAHL confirms
this assessment.(22)
Protective effect of different heating media
3.32 Mr Hickey asserted that AQIS had reviewed scientific literature
on heat inactivation of NDV and has taken any protective effects fully
into and determining the effectiveness of the proposed quarantine treatments.
He further noted that Dr Alexander's letter to Mr Baldwin indicated that
he (Alexander) doubted whether chicken nugget would differ significantly
from the homogenised bursae in terms of its protective effects on viruses.
3.33 The BRS study also refutes the suggestion that the use by Alexander
of bursal homogenate was an invalid method of testing thermostability.
The BRS noted that:
...the 50% bursa homogenate used would have contained natural buffering
components and would have approximated naturally contaminated tissues.
The virus used allowed the trial to be "closest to field conditions".(23)
Alexander's concerns
3.34 As noted in the preceding paragraphs, Alexander expressed two reservations
about AQIS's use of his data:
- extrapolation of data to temperatures that had not been tested; and
- probability of survival of virus and starting infectivity level
Extrapolation of data
3.35 Mr Hickey refuted suggestions that AQIS had improperly extrapolated
Alexander's data to untested temperatures. He noted that the BRS report
had made similar comments. However, AQIS had used "internationally
accepted methods for calculating thermal processing standards."(24)
Probability of survival of virus and starting infectivity level
3.36 Mr Hickey said that the levels of virus used by Dr Alexander in
his study were very high compared to the likely levels to be found in
a vaccinated, clinically healthy but nonetheless infected chicken. BRS
advice provided to AQIS confirms this assertion.(25)
3.37 The BRS report suggests that Alexander conducted his experiment
using considerably higher concentrations of virus than would be found
in either chickens infected with the virus or in meat contaminated with
the virus after processing. He explained that the amount of virus in the
material Alexander tested was:
...possibly 1000 fold higher than might be expected from chicken meat
collected from a viraemic bird contaminated during processing with faeces
or bursal fragments.(26)
3.38 Mr Hickey indicated that the Australian Animal Health Laboratory
had provided AQIS with further information which was consistent with that
of the BRS, advising that properly vaccinated birds secrete less virus.(27)
Committee's conclusions
3.39 The Committee, as non-scientists, do not consider that it is appropriate
or wise to arbitrate on the issue of whether the science underpinning
AQIS's proposal is sound.
3.40 AQIS has put forward a considerable amount of evidence supporting
the view that its science is sound. Reputable scientific opinions including
those of most of the state veterinary authorities, the Australian Animal
Health Laboratory, the Bureau of Resource Sciences and Dr Dennis Alexander,
a renowned international virologist support AQIS's view.
3.41 The Committee notes that Dr Alexander qualified his support with
a warning about the need for care when adapting scientific findings to
a commercial situation:
...some assessment of the closeness of the conditions in the laboratory
and the commercial processing should be made and the extent to which
variations are likely to affect the desired outcome.(28)
3.42 Equally, however, scientists within the Australian poultry industry
have raised areas of doubt about some of the assumptions AQIS has made.
The Committee is reluctant to dismiss these concerns lightly.
3.43 The major areas of doubt appear to be a lack of specific knowledge
about the behaviour of Newcastle disease virus in chicken meat and the
characteristics of different strains of both IBD and ND. While considerable
scientific opinion supports AQIS's assumptions that ND is more susceptible
to heat treatment than IBD, it is likely that this will remain as a point
of contention unless AQIS produces incontrovertible evidence.
3.44 The Committee recommends that AQIS commission tests of
the thermal inactivation of Newcastle disease virus and infectious bursal
disease virus in chicken meat products. This study should be conducted
in conditions that are as close as possible to commercial processing conditions.
3.45 The Committee further recommends that AQIS collaborate
with industry in the design of this experiment so that there can be no
further disputation about the validity of the science underlying the protocols.
Chapter 4
ENDNOTES
- Evidence, 13th September 1996, p. 204.
- Evidence, 20th September 1996, p. 24.
- Evidence, 28th August 1996, p. 21.
- Evidence, 13th September 1996, p. 136.
- Evidence, 13 September 1996, pp 170-3.
- Evidence, 13th September 1996, p. 129.
- Evidence, 13th September 1996, p. 135.
- Evidence, 13th September 1996, p. 129.
- Correspondence, MacKenzie, 23 October 1996
- Evidence, 13th September 1996, p. 136.
- Evidence, 13th September 1996, p. 137.
- Evidence, 13th September 1996, p. 137.
- Evidence, 13th September 1996, p. 161.
- Evidence, 13th September 1996, p. 172.
- Evidence, 13th September 1996, p. 171.
- Evidence, 13th September 1996, p. 171.
- Evidence, 30 September 1996, p. 214.
- Evidence, 30th September 1996, p. 219.
- Evidence, 30th September 1996, p. 219.
- Evidence, 30th September 1996, p. 215.
- Evidence, 30th September 1996, p. 232.
- Evidence, 30th September 1996, p. 218.
- AQIS position paper, op. cit., p.22.
- Evidence, 30 September 1996, p. 214.
- Evidence, 30 September 1996, p. 216-7.
- Gard, op. cit., p.11.
- Evidence, 30th September 1996, p. 217.
- Evidence, 13th September 1996, p. 171.