Content uploaded by Michael Graeme Garner
Author content
All content in this area was uploaded by Michael Graeme Garner on Mar 03, 2014
Content may be subject to copyright.
Epidemiol.
Infect.
(1998),
121,
409-418.
Printed
in
the United Kingdom
0
1998
Cambridge University Press
Viral haemorrhagic disease
of
rabbits and human health
J.
A. CARMAN'*, M.
G.
GARNER',
M.
G. CATTON3,
S.
THOMAS',
H.
A.
WESTBURY4,
R.
M.
CANNON',
B.
J.
COLLINS4
AND
I.
G.
TRIBE'
Adelaide, South Australia, 5000
Communicable Disease Control Branch, South Australian, Department
of
Human Services, PO Box 6,
Bureau
of
Resource Sciences, PO Box
Ell,
Kingston, ACT,
2604
Victorian Infectious Diseases Reference Laboratory, Fairfield Hospital, PO Box 65, FairJeld, Victoria,
3078
Australian Animal Health Laboratories, Private Bag
24,
Geelong, Victoria,
3213
(Accepted
6
April
1998)
SUMMARY
Viral haemorrhagic disease of rabbits (VHD), a potential biological control for wild rabbits in
Australia and New Zealand, escaped from quarantined field trials on Wardang Island and
spread to the mainland of Australia
in
October 1995. This study looked for any evidence of
infection or illness in people occupationally exposed to the virus. Two hundred and sixty-nine
people were interviewed and 259 blood samples were collected. Exposures to VHD-infected
rabbits ranged from nil to very high. No VHD antibodies were detected in any of the 259 sera
when tested by VHD competitive enzyme immunoassay, which had been validated with 1013
VHDV-specific antibody negative sera. A questionnaire designed to elicit symptoms of disease
in
a range of organ systems found
no
significant differences between illness in those exposed
and those not exposed to VHD, nor could an association be found between exposure and
subsequent episodes of illness. The findings are consistent with the view that exposure to VHD
is not associated with infection or disease in humans.
INTRODUCTION
The European rabbit
(Oryetolagus eunieulus)
was
introduced into Australia
in
the 1850s. Since then, it
has populated much of Australia and is now regarded
as a major agricultural and environmental pest [l]. In
1989, Australia and New Zealand began investigations
into the use of rabbit calicivirus (also known as viral
haemorrhagic disease virus), the cause of viral
haemorrhagic disease
of
rabbits (VHD), as a possible
biological control agent for wild rabbits. VHD was
first recognized in China in 1984 following the
introduction of rabbit stock from Germany [2]. It
spread rapidly in China, covering
50000
square
kilometres in less than 9 months and killing 470000
*
Author for correspondence.
rabbits
in
the first 6 months [3,4]. VHD subsequently
spread into Asia and Europe and to some American
and African countries, killing millions of rabbits [5].
VHD has now been reported in over
40
countries.
Viral haemorrhagic disease virus (VHDV) was
imported into the microbiologically secure Australian
Animal Health Laboratory (AAHL) in 1991, where
testing commenced into its efficacy for killing wild
rabbits and for its species specificity. AAHL found
no
evidence that VHDV could infect any species other
than the European rabbit, confirming work in other
countries [6-91. Following these promising laboratory
studies, field trials commenced
on
Wardang Island,
South Australia (SA) in March 1995. In September
1995, VHDV escaped from quarantine. Despite the
implementation of contingency plans to eradicate the
410
J.
A. Carman and others
disease, the virus continued to spread rapidly across
South Australia and was reported interstate by
December 1995. Because of the large areas involved
and the anticipated costs, plans to eradicate the
disease were abandoned.
Despite the escape, assessment of the deliberate
release of VHD continued under the Commonwealth
Biological
Control
Act
1984,
a process that requires
public consultation. One of the main concerns raised
was the perceived potential for VHDV to infect
species other than rabbits, including humans. While
there was
no
evidence in the scientific literature of
animals other than the European rabbit being
infected, there were also no reports of scientific studies
into the human health effects of the virus. Conse-
quently,
a
Human Health Study Group was formed to
plan and manage a study to determine if there were
any links between exposure to VHD and serological
and clinical evidence of infection with the virus in
humans.
METHODS
The study was undertaken in July 1996 and involved
serological testing and a questionnaire survey of
people with occupational exposure to VHDV. Advice
from overseas groups and laboratories working with
the virus was also sought.
Study subjects
Participants were mostly government employees
working in animal health laboratories, wildlife man-
agement or agriculture. Most exposures to VHD
occurred in field and laboratory staff involved in the
unsuccessful eradication campaign in South Australia
during October-November 1996, and those in contact
with wild rabbits in areas where VHD was active.
More recent exposures occurred in field staff located
in the south east of South Australia and in Victoria
where the disease was first reported in March 1996.
Government agencies assisted by providing lists of
staff. People were approached by telephone and asked
to participate in a study to investigate potential links
between exposure to wild animals and health. Par-
ticipation was voluntary. Refusals occurred on the
grounds of disliking blood sampling, travel time,
distance, or lack
of
interest. Trained nurses gave the
participants a written explanation of the study,
an
information sheet and obtained a signed consent form
before taking blood and completing a questionnaire.
In order to increase participation, nurses went
to
over
30
places throughout the
2
states and sampled
participants in their workplace or in various
laboratories.
The study was approved by the Ethics Committee
of
the Royal Adelaide Hospital of South Australia.
Serology
Approximately
20
ml of whole blood was collected
from each participant. In South Australia, sera were
separated at local laboratories and transported di-
rectly to AAHL for serological testing. Victorian
specimens were sent to Victorian Infectious Diseases
Reference Laboratory (VIDRL) for serum separation
and subsequent referral to AAHL.
A competitive enzyme immunoassay (EIA) for
VHDV-specific antibody, the most sensitive and
specific test for VHDV antibody in rabbit sera [lo],
was chosen as the screening test for the study. This
EIA uses purified whole VHDV particles as antigen,
and uses rabbit antiserum, hyperimmune to VHDV,
and normal rabbit serum as positive and negative
controls respectively. The assay was performed as
described
[
101. Results were expressed as percentage
inhibition values determined by the formula:
O/O
inhibition
=
100
x
[I-(OD test serum/OD negative
control)]. All sera were tested in duplicate and the
mean of the two values was recorded. Sera with a
percent inhibition of greater than
50%
were con-
sidered to be positive, between 30% and
50%
was
considered to be equivocal, and less than 30% was
considered to be negative to VHDV antibodies.
Testing was performed blind by laboratory staff.
Under the testing protocol (Fig.
1)
any reactors in
the competitive ETA were to be tested in a second
indirect EIA
[
101 to clarify their serostatus.
Prior to their use in the study, the specificity of both
EIAs was evaluated using a panel of 1000 metro-
politan blood and tissue donor sera predating the
presence of VHDV in Australia. An additional 13
human sera containing specific antibody to the human
viruses most closely related to VHDV, the Norwalk-
like group of viruses and hepatitis E virus, were also
tested to exclude cross reactivity attributable to these
antibodies. These sera were obtained from the VIDRL
reference serum collection where they had been
maintained at
-20
"C.
The competitive EIA was
evaluated using all 1013 samples of this panel of sera.
A more limited evaluation
of
the indirect EIA was
performed using
200
of these sera. Assay sensitivity
VHD of rabbits and human health
411
[
VHD
human testing protocol
I
Competitive
EIA
I
equivocal
\
-/
Report negative
i
[
Report indeterminate
Report positive
4]
I
Fig.
1.
VHD
human
testing
protocol.
for human serology could not be assessed, since no
humans were known to have been infected with
VHDV.
International survey
As part of this study, a range
of
international
laboratories and groups working with VHDV were
contacted. Information was sought about the
consequences of parenteral or mucous-membrane
exposure to VHDV-infected material, as well as any
evidence of human infections.
The questionnaire
A
questionnaire was used to collect information on
exposures to rabbits and health status, including
illnesses, over the period July 1995 to July 1996, as
well as alcohol intake, age, sex, occupation and level
of education. In order to reduce possible recall bias
(and to avoid unnecessary concern in those exposed to
VHDV), potential links between VHD and human
illness were not mentioned, questions on health were
asked before questions relating to exposure, and
participants were first asked about exposure to a
variety of Australian wild animals before being asked
more specific questions about exposure to rabbits.
The questionnaire sought symptoms associated
with hepatic, gastrointestinal and coagulation
disorders in view of the natural history of VHD in
rabbits
[8,
111 and the symptoms in humans of the
related Nonvalk-like and hepatitis
E
viruses. Because
of
the unknown nature
of
any infection in humans,
the range
of
symptoms under investigation was
extended by testing for possible viral effects on other
body systems. A set of
26
symptoms was used,
grouped into six illness categories (Table
4).
For each
bout
of
illness in the study period, the symptoms, the
onset date and period
of
illness was determined as
accurately as possible, often with the help of diaries.
Symptoms and illnesses associated with non-
infectious causes, such as surgery, diabetes or nausea
due to pregnancy and previously-occurring chronic or
previously-investigated problems such as long stand-
ing dermatitis, asthma or symptoms associated with a
previously diagnosed Ross River virus infection, were
excluded. Because alcohol consumption is associated
with hepatitis, alcohol consumption was also
412
J.
A. Carman and others
Table
1.
'Types
of
contact with rabbits and number
of
exposures
to
VHD
V
Type
of
contact with rabbits
Number
of
reports
of
contacts with rabbits
Number
of
people
with this as the
VBD VHD not Exposure to highest level
of
present present VHDV exposure
- -
No contact with rabbits
Nil
118
Ripped warrens, poisoned or fumigated
91
250
Low
43
rabbits; handled, cut open or skinned rabbits
with a plastic barrier e.g. gloves
bare hands
Handled, cut open or skinned rabbits with
115
23
1
High
108
206
48
1
2
69
measured as a potential confounder of any relation-
ship between exposure to VHDV and hepatitis.
Level
of
exposure
to
VHDV
For each person, exposure to VHDV was determined
by measuring as accurately as possible the time,
location, type
of
exposure to rabbits and the number
of rabbits contacted for each encounter during the
period July 1995 to July 1996.
A
series of detailed
maps, documenting the presence of VHD at various
times, were prepared from laboratory-confirmed
reports, held by the South Australian Animal and
Plant Control Commission
(J.
Kovaliski, personal
communication 1996). On the basis of these maps,
participants who reported that they had handled
rabbits in areas where VHD was known to have been
present at the time, were considered to have been
exposed.
The level of exposure (Table
1)
was defined in terms
of contact with rabbit body fluids, since VHDV
appears to be transmitted from one rabbit to another
by contact with excretions or body-fluids [12]. Thus,
exposure was classified as high for subjects who
reported skin contact with the body fluids of an
infected rabbit, and low for those who reported
contact with VHDV-infected rabbits, but did not have
contact with their body fluids. Shooting rabbits
without picking-up the remains was not considered to
be an exposure. Rabbits, when eaten, were all well-
cooked,
so
eating rabbits was also not considered to
be an exposure.
Data were analysed using direct analysis and the
statistical packages EpiInfo version 6.04
[I
31
and
SAS
version 6.03
(SAS
Institute Inc. Cary, NC, USA).
RESULTS
Questionnaires were completed by 269 people, 168
people from South Australia and 101 from Victoria.
Blood
could not be sampled from
10
of these, resulting
in
160 and 99 blood samples from each State
respectively.
Serology
There were two responses in the equivocal region
(percent inhibition
30-50
YO),
but no positive reactivity
(>
50%),
in the competitive
EIA
on testing the
evaluation panel of
1000
negative sera and the 13 sera
containing antibodies to Norwalk-like and hepatitis
E
viruses (Fig. 2).
No
reactivity was observed on testing
200 of these
1000
negative sera in the indirect
EIA
(not shown). Both assays were accordingly used
to
test
the 259 study sera without modification of the assay
protocols used for rabbit sera.
None of the 259 test sera were positive, or gave
reactivity in the equivocal region in the competitive
EIA (Fig.
3).
Moreover, the distribution of com-
petitive inhibition values for these sera was com-
parable to the 1013 presumed VHDV-negative sera. A
significant difference in percentage inhibition values
was noted between South Australian and Victorian
sera (Table 2, Fig.
3),
possibly reflecting variations in
processing between the states, prior
to
testing. There
was no relationship between exposure status and
percentage inhibition values for either set of sera
(Table 2).
Although the absence of any reactors obviated the
need for supplementary testing, all 259 test sera were
also tested using the indirect
EIA
in case factors
unique to the competitive
EIA
format had contributed
to false negativity. All samples tested negative. In
VHD
of
rabbits and human health
413
80
70
60
50
c
$
40
2
30
20
10
0
LI.
-
)
-16 -12
-8
4
0
4
8
12 16 20 24 28 32 36 40
Percent inhibition
Fig.
2.
Frequency distribution of the serological results by competitive EIA for the
1000
negative sera and the
13
sera that
were positive for Norwalk-like antibodies and hepatitis E antibodies.
25
20
6
15
B
g
10
$
5
0
-6
4
-2
0
2
4 6
8
10 12 14 16
18
20 22 24
Percent
inhibition
Fig.
3.
Frequency distribution of the
269
test sera by competitive
EIA.
Table
2.
Serological
results
Difference between
exposure categories
by ANOVA
(P
values)
South Australia Victoria
Exposure Within Within Between
to VHD Number Mean
S.D.
Number Mean
S.D.
SA Victoria states
Competitive EIA None
67
11.12
4.45
44 3.91 3-93
Percent inhibition Low
35 10.60 403
5
7.20
460
High
58 10.53
4.99
50
4.82 3.82
Indirect EIA None
67 0.06
0.02
44
0.05
0.02
Optical density Low
35
0.07
0.03
5
0.06
0.06
High
58
0.07
0.03
50
0.05
0.03
075
0.10
<
0.001
0.16
0.53
0.00
1
addition, optical density values did not relate to
exposure status (Table
2).
While
a
scattergram
of
competitive
EIA
and indirect
EIA
results (not shown)
showed a linear relationship between the two, the
highest values for each test were obtained by different
sera.
414
J.
&4.
Carman and others
Table 3.
Number
of
episodes
of
illness for all respondents and date ofjirst exposure to
VHD
by month
Number
of
respondents
with
Episodes
of
illness
first exposure
to
Month
of
~
VHD exposure Any Diarrhoea Fever Rash
Bleeding
Hepatitis Neurological
38*
Jul
1995
24 2 23
0
0
0
3
3
Aug
23 3
20 1
0
0
3
2
SeP
15 7
14
1
1
0
3
52
Oct
23
10
20
0 0
0
3
9
Nov
19
11
11
1
0
0
4
3
Dec
10 6
7 1
0 0
3
7
Jan
1996
15
6
13
0
1
0
4
8
Feb
14
6
12
0
0
0
3
6
Mar
21 9 20
0
0
0
5
7
APr
35 17 30 2
0 0
8
5
May
29 11 25
0
1
0
6
10
Jun
64 18
57 7
1
1
15
1
Jul
42 10
40
1
0
1 9
Unknown
I1
6
10 2
1
1 2
151
Total
345 122 302 16 5
3
71
*
Including
or
before
July
1995.
International survey
Responses were received from
47
groups in 16
countries. There were no reports of VHDV trans-
mission to humans despite prolonged or repeated
exposure to material containing high titres of VHDV.
These exposures included a worker with multiple
splashes of infected rabbit blood in the mouth and
three descriptions of VHDV-contaminated sharps
injuries-on multiple occasions in two
cases.
No
subsequent ill effects were reported. Three respondents
had undertaken serological testing of small numbers
of their laboratory staff, two respondents on more
than one occasion.
No
positive results were recorded.
Questionnaire results
Of the 269 people interviewed,
11
8
had no exposure,
43 had low exposure and
108
had high exposure to
VHDV. The latter group included
19
people with
broken skin on their hands who had cut open or
skinned infected rabbits. Rangers, farmers, agricul-
tural advisers and veterinary officers had the highest
levels of exposure. Although there was no difference
in age
or
educational level between those exposed and
unexposed, there was a significant sex difference
(&,
=
33,
P
=
<
O.OOl),
with more males than fema-
les exposed to VHDV. There was a significant
difference
in
alcohol intake between the exposed and
unexposed groups
(x,"z,
=
8.07,
P
=
0.02), but this
difference disappeared when the study population was
stratified by sex.
Table 3 shows that the distribution of reported
episodes of symptoms by month was not uniform,
with a peak in the winter months of June and July,
1996. Table
4
also shows the distribution by month of
first exposure to VHD in participants. The exposure
figure for July 1995 includes exposures before that
date, and the large peak in October 1995 corresponds
to the exposures during the eradication campaign
following the escape of the virus from quarantine. The
average period of observation after first exposure was
8.1
months, although disease information was col-
lected from each participant for the full
13
months.
If VHDV causes illness, a higher rate of symptoms
and illness in the exposed groups would be expected,
regardless of when the participants were exposed,
Accordingly, the first analysis compared illness over
the entire 13 month study period for the three groups
of exposure (nil, low and high). The 26 symptoms and
their frequency of occurrence were put into six specific
illness groups, and one non-specific group (Table
4).
The number of episodes of illness were first compared
between the unexposed and combined exposed
categories. Thenumber ofepisodes in the threeexposure
categories were then compared. There were no
significant differences in the individual symptoms
between the exposure categories, except for 'presence
of dark urine' and 'difficulty speaking and seeing'. All
the cases of the first symptom occurred in the
VHD
of
rabbits
and
human health
415
Table
4.
Reported illnesses and symptoms from
July
1995
to
July
1996,
by
exposure category
Exposure level to VHD
Illness
P
values
Respondents reporting illness Nil Low High
Episodes reported Per
100
respondents Nil
vs.
Symptoms reported Nil Low High Nil
vs.
low
vs.
Number
of
respondents
118 43
108
100.0
100.0
100.0
exposed high
Diarrhoealgastro-intestinal
illness
Respondents reporting illness
Episodes reported
Diarrhoea
Vomiting
Nausea
Stomach cramps
Blood in stools
Fever/’flu-like symptoms
Respondents reporting illness
Episodes reported
Fever
Sweats or chills
Unusual tiredness or
‘
off
colour
’
Swollen glands
Flu-like illness
Sore red or weepy eyes
Sore joints or muscles
Rasheslskin conditions
Respondents reporting illness
Episodes reported
Unusual rashes or blisters
on
skin
Bleeding problems
Respondents reporting illness
Episodes reported
Unusual bruising
Unusual bleeding
Hepatitis symptoms
Respondents reporting
Episodes reported
Yellow skin or eyes
Neurological symptoms
Respondents reporting illness
Episodes reported
Dizziness
Blacking out
Incoordination, difficulty walking
Difficulty speaking or seeing
Fits
Weakness or tingling in limbs
Non-specific symptoms
Dark urine
Pale stools
Loss
of
appetite
Headache
Respondents reporting illness
Episodes reported
Any illness
42
60
45
13
31
31
5
83
145
75
70
120
49
119
21
65
5
7
7
1
1
0
1
1
1
1
24
31
28
1
8
0
0
4
12
3
64
94
90
166
14
17
13
2
9
10
1
27
45
30
24
33
13
34
3
15
2
2
2
0
0
0
0
1
1
1
9
15
12
0
3
5
0
1
0
0
17
26
30
52
38
45
34
17
31
19
2
80
112
63
62
91
31
86
16
47
6
6
6
2
3
0
3
1
1
1
21
28
25
1
6
2
0
5
0
1
42
74
82
127
35.6
50.8
38.1
11.0
26.3
26.3
4.2
70.3
122.9
63.6
59-3
101.7
41.5
100.8
178
55.1
4.2
5.9
5.9
0.8
0.8
0.0
0.8
0.8
0.8
0.8
20.3
263
23.7
0.8
6.8
0.0
00
3.4
10.2
2.5
54.2
79.7
76.3
140.7
32.6
39.5
30.2
4.7
20.9
23.3
2.3
62.8
104.7
69.8
55.8
76.7
30.2
79
I
7.0
34.9
4.7
4.7
4.7
0.0
0.0
0.0
0.0
2.3
2.3
2.3
209
34.9
27.9
0.0
7.0
11.6
0.0
2.3
0.0
0.0
39.5
60.5
698
120.9
35.2
0.87
41.7
0.24
31.5
0.33
15.7
0.71
28.7
0.97
176
0.22
1.9
0.29
74.1
0.96
103.7
0.15
583
0.84
574
0.80
84.3
0.09
28.7
0.09
79.6
0.06
14.8
0.27
43.5
0.10
5.6
0.69
5.6
0.83
5.6
0.83
1.9
0.71
2.8
0.45
0.0
n/a
28
0.45
0.9
0.71
0.9
0.71
09
0.71
19.4
0.93
25.9
0.73
23
1
0.90
09
0.86
5.6
0.79
1.9
0.02
0.0
n/a
4.6
0.81
0.0 0.00
0.9
0.21
38.9
0.07
68.5
0.20
759
0.84
I176
0.11
0.96
0.49
0.62
0.19
0.70
0.38
0.56
0.76
0.35
0.7 1
0.96
0.22
0.23
0.18
0.29
0.20
0.90
0.95
0.95
0.58
0.34
0.34
n/a
0.7 1
0.7 1
0.7 1
0.98
0.60
0.86
0.82
0.92
0.0002
0.78
n/a
0.0005
0.42
0.19
0.38
0.9 1
0.28
416
J.
A. Carman and others
Table
5.
Expected episodes
of
illness and exposure to
VHD,
taking into uccount different periods
of
observation
in
the exposed and unexposed groups
Exposed
(number)
Total Non-exposed
P
value
Non-exposed Exposed
(number)
(number)
Observed Expected
(x&)
(standardized) (standardized)
Person months
of
3497 2129 1368
100.0
1000
Number
of
episodes
observation
Any
334 204
130 130.66 0.94
9.58
950
Diarrhoea
116 67
49 45.38
0.49
3-15 3.58
Fever
292 177
115
1
14.23 0.92
8.3
1
8.4
I
Rash
14 7
7 5.48 0.40
0.33
0.51
Hepatitis
2
0
2 0.78
n/a
0.00
0.15
Neurological
69 36
33 26.99 0.14
1.69
2.4
1
Bleeding
4 1 3 1.56
n/a
0.05
0.22
unexposed group. All the cases of the second symptom
came from the two exposed categories, but three
episodes were from the same person, with the first of
these episodes occurring
2
months before that person’s
first low level exposure to VHDV. After discounting
this person, there was no significant difference between
the groups.
There were also no significant differences in the
numbers of different illnesses reported nor in the
numbers of people reporting illnesses, in the three
exposure categories. For those respondents reporting
episodes of illness, the nil exposure category ex-
perienced an average of 1.85 episodes of illness per
person, compared to 1.73 episodes per person and
1.54
episodes per person in the low and high exposure
categories respectively.
A further analysis was done to see
if
there were
more illnesses immediately following exposure. We
would have liked to have considered the number
of
illnesses in the
30
days following exposure to VHDV,
but the dates obtained were not precise enough for
such an analysis. Indeed 11 episodes of illness had
insufficient date information and had to be excluded.
Each bout of illness was allocated to the exposed
group if the respondent had been exposed to VHDV
in that or a previous month, otherwise it was allocated
to the unexposed group. Table
5
shows that the period
of observation for exposed and non-exposed respon-
dents was 1368 person months (39.1%) and 2129
person months (60.9
YO)
respectively. An analysis that
took these proportions into account found no sign-
ificant difference in the number of episodes of illness
between the exposed and non-exposed groups (Table
5).
Because seasonal variation of disease over the study
period (as noted in Table 3) can cause a significant
result in such an analysis, episodes of illness were also
examined more closely with a series of contingency
tables on a month-by-month basis. For each month,
respondents were categorized in two ways
:
whether
they were exposed to VHD that month and whether
they reported symptoms of illness in the period of that
and the following month. This eliminates any seasonal
effect, provides a follow-up period of
4-8
weeks after
reported exposure and increases the sensitivity of
finding an association between exposure and illness. If
there was an effect, an increase in the amount of illness
over the 2-month period would be expected. However
there would also be a correlation between the
successive monthly analyses because the correspond-
ing 2-month periods for counting illnesses overlap.
No
association between exposure to VHDV and
illness in humans was found.
DISCUSSION
This study looked for links between occupational
exposure to VHDV-infected rabbits and illness in
humans. The study had two parts, a serological survey
of the study participants to determine if antibodies to
VHDV could be detected and an epidemiological
investigation into links between exposure to VHD and
clinical symptoms of illness.
VHDV-specific antibody was selected as a marker
of VHDV infection because it is a persistent marker of
past infection in rabbit survivors of VHD. Rabbits
maintain detectable levels
of
antibodies for at least
130 days after infection [lo]. Infections with human
VHD of rabbits and human health
417
caliciviruses also give rise to long-lasting serological
reactions [14].
No
VHDV-specific antibodies were detected in any
sera using either of the competitive EIA or the indirect
EIA. The competitive inhibition values and optical
density readings were all well below the respective
positive assay cut-offs, and below the assay equivocal
zones. These assay thresholds were developed using
rabbit sera, and in the absence of sera from VHDV-
infected humans, their validity from human serology
cannot be proven. There was, however,
no
relation-
ship observed between VHDV exposure status and
serological results, and the competitive inhibition
values from the 259 test sera fell into a homogeneous
normal distribution, comparable to that of the 1013
control sera used for essay validation. There was,
therefore,
no
evidence of a potential positive popu-
lation of sera poorly differentiated from negatives by
the assays employed.
These results support the findings of the survey of
groups working with VHDV in other countries.
No
respondent reported any case of human infection with
VHDV and there were
no
reports of illness associated
with exposure to VHD. Prior to this survey, the only
known incident of potential human transmission, was
an unpublished report of a low, transient antibody
response to VHDV in one person in Mexico with a
high level of exposure. The person did not get ill
(H.
C.
Lopez, Director General of Animal Health,
Mexico, letter to J.
G.
Murray 24 January 1996).
Given our findings, the antibody response is both
unlikely and unexpected and further information on
the test used, its sensitivity, specificity and
repeatability, would be required before the possibility
of
a test artefact could be eliminated.
For the epidemiological part of the study, it was
difficult to determine how to define exposure to a virus
that was not known to transmit to humans and how
to measure any resultant clinical illness, when illness
in humans had never been described. Exposure
categories were assigned
on
evidence that transmission
between rabbits occurs
by
contact with body fluids
[
121. Potential illness was assessed by considering a
range of symptoms, including general symptoms
associated with viral infections, symptoms observed in
rabbits with VHD, and symptoms of humans infected
with related viruses. A variety of analyses were done
with these data, comparing illness to exposure over
the entire study period and on a month-by-month
basis.
No
increase was found in the amount of illness
reported in exposed compared to non-exposed people.
A recent review paper
on
caliciviruses
[
151, suggests
that VHDV may pose a possible threat to humans,
citing as evidence data from the initial limited-
distribution report
of
this study
[16].
However, the
authors’ use of the data is misleading. They did not
mention the conclusions of the study and chose to
classify exposure
on
the amount of virus in the
environment. They noted that the number
of
cases of
illness in the period July-December 1995 was less than
in the period February-July 1996 (see Table
3),
and
argued that during the first period there was a low
level of virus, and during the second period, because
of the spread of the disease, a high level
of
virus. They
did not consider the actual exposure level of the
respondents, and more importantly, that many of the
respondents in the earlier period were exposed during
the initial attempts to contain the disease (as shown in
Table
3).
Since the initial report [16], the data have
been thoroughly rechecked, the results given here
differ slightly, but the conclusion remains the same
-
there were no significant difference between levels or
types of illness in those exposed and those not exposed
to VHDV.
In this type of study, there is always the potential
that clinical interviews may not detect sub-clinical
infections and that the brevity of the interview and its
timing, months after some exposures, may diminish
the likelihood
of
identifying minor symptoms. While
the absence of detectable antibodies in exposed
humans strongly suggests that infection has not taken
place,
no
information is available about whether
antibodies may be produced in infected humans
anyway, nor the duration of any such response.
Without human sera containing VHDV-specific anti-
body to verify the sensitivity of the assay, it is not
possible to definitively exclude the presence of VHDV
antibody in sera. In addition, the small sample size in
the study make it difficult to exclude the possibility of
rare or infrequent infections. These limitations not-
withstanding, the study presents a range of data
consistent with evidence from other countries, that
VHDV is not associated with infection or disease in
humans.
ACKNOWLEDGEMENTS
We wish to thank the following nurses who assisted
with collection of blood samples and interview data;
Maura Pottril, Sarah Dugdale, Adriana Milazzo and
Anne Murphy. The assistance
of
Jennie Leydon, Alan
Breschkin and John Marshall
of
VIDRL in preparing
418
J.
A.
Carman and others
the evaluation panel of human sera is acknowledged.
We would also like to thank Kieran McCaul for
statistical advice, David Roder for epidemiological
advice and David Creeper for general assistance. The
staff from the Communicable Disease Control
Branch, South Australian Department of Human
Services and the Victorian Department of Human
Services are also gratefully acknowledged, as is the
support
of'
the various agencies and their employees
who agreed to participate in the study. The intellectual
input of the other members of the RCV Human
Health Study group, being Cathy Mead, John Kaldor
and Scott Crerar, is also gratefully acknowledged. The
study was funded by the Bureau
of
Resource Sciences,
Department of Primary Industries and Energy,
Canberra.
REFERENCES
1. Williams CK, Parer I, Coman BJ, Burley
J,
Braysher
ML. Managing vertebrate pests: Rabbits. Bureau of
Resource Sciences and CSIRO Division of Wildlife and
Ecology, Australian Government Publishing Service,
Canberra, 1995.
2.
Liu SJ, Xu HP, Pu PQ, Quian NH. A new viral disease
in rabbits. Animal Husbandry Vet Med 1984;
16:
3. Gregg DA, House C, Meyer R, Berninger M. Viral
haemorrhagic disease of rabbits in Mexico
;
epidemi-
ology and viral characterization. Rev Scient Techn
Off
Intern Epizoot 1991
;
10:
435-51.
4. Xu
W.
.Viral haemorrhagic disease of rabbits in the
People's Republic of China
:
epidemiology and virus
characterization. Rev Scient Tech
Off
Intern Epizoot
1991;lO: 393408.
5.
Rodak
L,
Smid B, Valicek
L.
Application of control
measures against viral haemorrhagic disease of rabbits
253-25
5.
in the Czech and Slovak Federal Republic. Rev Scient
Techn
Off
Intern Epizoot 1991;
10:
513-24.
6.
Fan YY, Yu
M,
Yang HP, et al. The study of domestic
rabbit haemorrhagic virus (in Chinese, translated by D.
Sun
and L.Chang). Shanghai Dom Anim Hosp
Corresp/Commun 1987;
4:
7-9.
7. Wang Y, Ji C, Zhou Y, Sun H, et al. Study on a new
virus infection in rabbits
-
rabbit pest (in Chinese,
English abstract). Scient Agricult Sinica 1988;
21:
73-9.
8.
Xu ZJ, Chen WX. Viral haemorrhagic disease in
rabbits: a review. Vet Res Commun 1989;
13:
205-12.
9. Bureau of Resource Sciences. Rabbit calicivirus disease
:
a report under the Biological Control Act 1984. Bureau
of
Resource Sciences, 1996, Canberra.
10. Collins BJ, White JR, Lenghaus C, Boyd V, Westbury
H.
A competition EIA for the detection
of
antibodies to
rabbit haemorrhagic disease virus. Vet Microb 1995;
11. Marcato PS, Benazzi C, Vecchi
E,
Galeotti M, Della
Salda L, Sarli
G,
Lucidi P. Clinical and pathological
features of viral haemorrhagic disease of rabbits and the
European brown hare syndrome. Rev Scient Techn
Off
Intern Epizoot 1991;
10:
371-92.
12. Cancellotti FM, Renzi M. Epidemiology and current
situation of viral haemorrhagic disease
of
rabbits and
the European brown hare syndrome. Rev Scient Tech
Off
Intern Epizoot 1991;
10:
409-22.
13. Dean
AG,
Dean JA, Burton AH, Dicker RC. Epi Info,
Version 6.03
:
a Word processing, database and statistics
program for epidemiology on microcomputers. USD
Incorporated, Stone Mountain, Georgia, 1996.
14. Kapikan AZ, Greenberg HB, Cline WL, et al. Preva-
lence of antibody to the Norwalk agent by a newly
developed immune adherence haemaaglutination assay.
J Med
Virol
1978;
2:
281-94.
15. Smith AW, Skilling DE, Cherry
N,
Mead JH, Matson
DO.
Calicivirus emergence from Ocean reservoirs
:
zoonotic and interspecies movements. Emerg Infect Dis
16. Human Health Study Group. Rabbit calicivirus and
human health. Report of the Rabbit Calicivirus Human
Health Study Group, 1996.
43:
85-96.
1998;
4:
13-20.
