An investigation into human pandemic influenza virus (H1N1) 2009 on an Alberta swine farm

Abstract

On May 2, 2009 the Canadian Food Inspection Agency notified the World Organization for Animal Health that an emerging novel influenza A virus (pandemic H1N1 2009) had been confirmed on a swine farm in Alberta. Over a 4-week period pigs in this farrow-to-finish operation were clinically affected by respiratory disease consistent with an influenza A virus infection and the presence of active viral infection was confirmed in all production areas by real-time polymerase chain reaction (RT-PCR). Despite clinical recovery of animals, there was reluctance by purchasers to receive animals from this operation due to concerns about the effect on both domestic and international markets. The owner decided to depopulate the entire herd due to impending welfare issues associated with overcrowding and economic concerns resulting from the inability to market these animals. Carcasses were rendered or composted and did not enter the human food or animal feed chain. The source of virus in this herd was determined to be an infected human. Zoonotic transmission to 2 individuals responding to the outbreak was suspected and recommendations to prevent occupational exposure are discussed.

Full text

CVJ / VOL 50 / NOVEMBE R 2009 1153
Special Report Rapport spécial
Terrestrial Animal Health Division, Canadian Food Inspection Agency, Ottawa, Ontario K1A 0Y9 (Howden, Caya, Morrison,
Laurendeau); Production Animal Health, University of Calgary, Faculty of Veterinary Medicine, 3330 Hospital Drive NW,
Calgary, Alberta T2N 4N1 (Brockhoff); Alberta Health Services, Central Zone, 300 Jordan Parkway, Red Deer, Alberta T4P 0G8
(McLeod, Lavoie, Ing); Department of Community Health Sciences, Faculty of Medicine, University of Calgary, 3330 Hospital
Drive NW, Calgary, Alberta T2N 4N1 (McLeod); Food Safety Division, Alberta Agriculture and Rural Development, Edmonton,
Alberta T6H 4P2 (Bystrom, Keenliside); National Centre for Foreign Animal Disease, 1015 Arlington Street Winnipeg, Manitoba
R3E 3M4 (Alexandersen, Pasick, Berhame); Operations Strategy and Delivery Directorate, Canadian Food Inspection Agency,
1400 Merivale Road, Ottawa, Ontario K1A 0Y9 (Rohonczy).
Address all correspondence to Dr. Krista Howden; e-mail: Krista.Howden@inspection.gc.ca
Reprints will not be available from the authors.
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA
office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
An investigation into human pandemic influenza virus (H1N1) 2009
on an Alberta swine farm
Krista J. Howden, Egan J. Brockhoff, Francois D. Caya, Laura J. McLeod, Martin Lavoie, Joan D. Ing,
Janet M. Bystrom, Soren Alexandersen, John M. Pasick, Yohannes Berhane, Margaret E. Morrison,
Julia M. Keenliside, Sonja Laurendeau, Elizabeth B. Rohonczy
Abstract — On May 2, 2009 the Canadian Food Inspection Agency notified the World Organization for Animal
Health that an emerging novel influenza A virus (pandemic H1N1 2009) had been confirmed on a swine farm in
Alberta. Over a 4-week period pigs in this farrow-to-finish operation were clinically affected by respiratory disease
consistent with an influenza A virus infection and the presence of active viral infection was confirmed in all pro-
duction areas by real-time polymerase chain reaction (RT-PCR). Despite clinical recovery of animals, there was
reluctance by purchasers to receive animals from this operation due to concerns about the effect on both domestic
and international markets. The owner decided to depopulate the entire herd due to impending welfare issues
associated with overcrowding and economic concerns resulting from the inability to market these animals. Carcasses
were rendered or composted and did not enter the human food or animal feed chain. The source of virus in this
herd was determined to be an infected human. Zoonotic transmission to 2 individuals responding to the outbreak
was suspected and recommendations to prevent occupational exposure are discussed.
Résumé — Enquête sur le virus de l’influenza pandémique humaine (H1N1) en 2009 dans une ferme porcine
de l’Alberta. Le 2 mai 2009, l’Agence canadienne d’inspection des aliments a informé l’Organisation mondiale
de la santé animale qu’un nouveau virus émergent de l’influenza A (H1N1 pandémique 2009) avait été confirmé
dans une ferme porcine en Alberta. Pendant une période de 4 semaines, les porcs de cette exploitation de la
parturition à la finition ont été cliniquement affectés par une maladie respiratoire présentant des symptômes
conformes à l’infection du virus de l’influenza A et la présence d’une infection virale active a été confirmée dans
toutes les aires de production par une réaction d’amplification en chaîne par la polymérase en temps réel. Malgré
le rétablissement clinique des animaux, il y avait une réticence de la part des acheteurs à recevoir les animaux de
cette exploitation en raison de préoccupations à propos de l’effet sur les marchés intérieurs et internationaux. Le
propriétaire a décidé de dépeupler le troupeau au complet en raison d’enjeux imminents liés au bien-être découlant
du surpeuplement et de préoccupations économiques découlant de l’incapacité de vendre ces animaux sur le marché.
Les carcasses ont été équarries et n’ont pas accédé à la chaîne alimentaire humaine ou animale. La source du virus
dans ce troupeau a été déterminée comme étant un humain infecté. La transmission zoonotique à 2 intervenants
lors de l’éclosion a été soupçonnée et des recommandations pour prévenir l’exposition au travail sont discutées.
(Traduit par Isabelle Vallières)
Can Vet J 2009;50:1153–1161

1154 CVJ / VOL 50 / NOVEMBER 2009
RAPPORT SPÉCIAL
Introduction
Influenza virus infections of swine occur commonly worldwide
on a year-round basis (1). Swine Influenza Virus (SIV) is a
synergistic factor in the porcine respiratory disease complex
(PRDC) and is an important cause of broncho-interstitial
pneumonia and respiratory disease in pigs. Influenza virus is a
zoonotic agent of concern on a global scale presenting economic
and health challenges to human and animal populations (2).
Although human H3N2 viruses have been isolated from pigs
in Asia and Europe, historically there has been varied evidence
of human H1N1 influenza viruses maintaining themselves in
swine populations (3). Both H3N2 and H1N1 SIV infections
have been reported in humans in Canada, the United States,
Europe, and Asia (4). Humans occupationally exposed to pigs
are at increased risk for sero-conversion and for influenza-like
illness (ILI) attributable to SIV (5). The reported number of
SIV infections in humans, however, is negligible compared
to the number of people exposed to pigs (6). The true inci-
dence and significance of zoonotic swine influenza infection is
unknown, in part due to inconsistent diagnostic confirmation
and reporting within and between jurisdictions.
On April 28, 2009 the owner of a conventional 220-sow
single site commercial farrow-to-finish swine operation in
Alberta notified his herd veterinarian of an acute onset cough
in his pre-grower and grower animals (Figure 1: D and E).
A contract worker hired to rebuild the ventilation inlets and
upgrade the exhaust fans in areas D and E had recently returned
from Mexico and exhibited symptoms of ILI while working in
the barn. Concerned about a potential public health risk the
producer reported these findings to his herd veterinarian, who
notified Alberta Agriculture and Rural Development (ARD) of
this situation.
Swine influenza is not a reportable disease under the federal
Health of Animals Act and there is no national control program
for this disease. However, authority is granted to the Canadian
Food Inspection Agency (CFIA) under this Act to respond
to any disease of animals including emerging diseases that
may affect animals or that are zoonotic. There are also certain
provinces, including Alberta, for which SIV is notifiable under
provincial animal health regulations. A notifiable disease dif-
fers from a reportable one in that occurrences of a notifiable
disease are recorded for surveillance purposes only and there is
no government response to confirmed cases (no quarantine or
disease control requirements) in Alberta. Based on the history
of ILI in both the humans and swine associated with this opera-
tion, public concern, and the scientific uncertainty surrounding
this emerging disease at the time, a joint decision was made by
CFIA and ARD for CFIA to issue a precautionary quarantine
under the Health of Animals Act for this herd and conduct a full
epidemiological investigation.
Health history of herd
Prior to this disease investigation the health status of this
swine operation was considered to be conventional stable.
The herd was porcine reproductive and respiratory syndrome
(PRRS) positive, Mycoplasma hyopneumoniae positive (MH), and
Actinobacillus pleuropneumoniae (APP) 5b positive. Although
PRRS positive for many years, the herd was stable for this virus.
Performance of the reproductive herd was above industry average
for farrowing rate and pigs weaned per sow per year. Routine
postmortem examinations and bacteriology completed during
regular herd visits revealed no gross pathology or bacterial iso-
lates to suggest that APP was active within the herd. Prior to
targeted vaccination of nursery pigs at 6 wk of age, MH had
been clinically active in area E feeder pigs (Table 1). Barn E is
the most challenged space with respect to air quality (dust, ven-
tilation rate). Vaccination against MH (Respisure-One/ER Bac
Plus; Pfizer Animal Health, Kirkland, Quebec) was routine at
6 wk of age and, through regular health and necropsy monitor-
ing, vaccination had successfully reduced losses and pathology
associated with this disease. Marketing records show that this
was a high index herd with an excellent health check record.
In November 2004, shortly after the introduction of pur-
chased gilts, there was a positive test on serology using the
swine influenza (H1N1) IDEXX antibody enzyme-linked
immunosorbent assay (ELISA). The parent herd of the gilts
had experienced an acute episode of SIV-related illness that
summer. Prior to shipping, these gilts had tested negative for
SIV on ELISA but had begun showing ILI shortly after arrival.
Subsequently, the breeding herd was vaccinated twice, 2 wk
apart, with an inactivated swine influenza virus type A, sub-
type H1N1 vaccine (Maxivac H1N1; Schering-Plough Animal
Health, Kirkland, Quebec). This vaccination strategy appeared
to have resolved the clinical concerns attributable to SIV. In the
subsequent 4.5 y, vaccination for SIV was not practiced and no
further SIV cases were identified. Thus it is unlikely that any
animals still present on this farm had been previously exposed
to, or vaccinated for, influenza.
Barn design and pig flow
Barn design and pig flow are consistent with a conventional
farrow-to-finish site. The 7 production areas (Figure 1) are all
connected by short hallways or doorways; no truly separate air-
spaces exist within this facility. The main entrance (Aa) houses
a boot change area only; there is no shower. Biosecurity for
visitors consists of changing boots and donning cloth coveralls
in an outbuilding prior to entering the main entrance. There
is no quarantine associated with this site, purchased breeding
stock (boars and gilts) are placed directly into Ac with imme-
diate proximity to the gestating and breeding herd (Ad, Ab).
Farrowing occurs in the newest area on site (B). Piglets are
weaned at 28 days and then moved into 1 of 6 all-in-all-out
(AIAO) hot nursery rooms (C). Pig flow from areas D through
G is continuous (Figure 1).
Regulatory response
Upon laboratory confirmation of the emerging pandemic H1N1
2009 virus within this herd, public and animal health authorities
at all levels became engaged in the development of a disease con-
trol strategy based on precautionary principles to address public
health concerns. Animal and public health authorities supported
the continuation of movement restrictions on this herd while
there was evidence of live virus circulating to prevent spread of

CVJ / VOL 50 / NOVEMBE R 2009 1155
S P E C I A L R E PO R T
disease to human or animal populations, pending assessment of
the behavior of the novel virus in swine populations. None of
the authorities supported a policy of eradication of the entire pig
herd on the basis of human or animal health risks. Additional
sampling and regular health assessments provided information
on the clinical course of disease and risk associated with pan-
demic H1N1 2009 virus in this herd.
With the entire herd under quarantine, finished hogs could
not be shipped to slaughter and crowding became an animal
welfare issue. To prevent distress in the animals and at the
request of the producer, ARD undertook a limited cull of
475 grower/finisher animals on May 8th to alleviate animal
welfare concerns and to allow time for repeat testing of the
herd. Hogs were humanely destroyed on-site by captive bolt
(“Cash Special” captive .22 and .25 caliber bolt stunner and
“Cash Special” HD captive .25 calibre bolt stunner) by trained
staff from ARD and the Alberta Society for the Prevention of
Cruelty to Animals. Staff were trained on the use of the captive
bolt stunners and veterinarians were present on-site at all times
during the depopulation activities.
Carcasses were transported to a rendering establishment for
disposal in an enclosed, leak-proof, fully covered conveyance via
a pre-planned route to minimize exposure to human and pig
populations. Routine biocontainment procedures, supervised
by the CFIA, were followed for movement of a conveyance off
an infected farm. Feeds division of the CFIA confirmed that
rendered carcasses could be used in animal feeds as influenza A
virus is heat labile and the time-temperature combination of the
rendering process would inactivate any virus. Despite scientific
evidence supporting the negligible risk associated with rendered
product, the rendered material was buried in a landfill due to
concerns by the rendering company about potential negative
public perception and the marketability of the meat and bone
meal produced. Disposal by burial was not an option because
the large volume of carcasses could not be accommodated on
the small farm property, there were concerns about public
perception, and soil type precluded the use of other burial sites
near the farm.
A controlled marketing approach allowing movement of test
negative animals to slaughter was proposed, with culling only to
address humane issues associated with overcrowding. The herd
was monitored by both the CFIA staff and the private veteri-
narian, and sampling was completed in each production area to
establish prevalence estimates and determine the clinical presen-
tation within each population of animals. In consultation with
stakeholders, criteria were developed to determine the clinical,
Figure 1. Barn layout and design to scale.
Table 1. Production areas and associated animal demographics
Production Age range
area Population Description (weeks) Flow
A 175 Breeding and
gestation sows
24 Gilts
2 Boars
B 48 Lactation sows
480 Nursing piglets 1–4 AIAO
C 600 Nursery 4–10 AIAO
D 340 Pre-grower 10–14 Continuous
E 420 Grower 14–18.5 Continuous
F 440 Grower/Finisher 18.5–21 Continuous
G 300 Finisher 21–25 Continuous

1156 CVJ / VOL 50 / NOVEMBER 2009
RAPPORT SPÉCIAL
laboratory, and epidemiological data needed to determine when
the disease was no longer present on this farm and the time at
which the quarantine could be released.
In spite of the uneventful clinical recovery of animals, no
slaughter facility would accept animals from this farm after the
quarantine was removed. Unfounded concerns about food safety
and marketability from meat buyers were cited as the reason
pork processors refused pigs from this farm. Due to impending
overcrowding in the barn, the herd owner made an economic
decision to depopulate the herd to allow him to escape the
situation and resume operation with a replacement herd. The
culling of the herd was not an ordered destruction by either the
CFIA or ARD on the grounds of animal or human disease con-
cerns. At the owner’s request, and with the assistance of ARD,
approximately 3000 pigs were humanely destroyed between
June 4th and 6th and either composted off site or disposed of
via rendering. The quarantine was removed on July 29, 2009
when cleaning and disinfection measures developed by the pri-
vate veterinarian and approved by CFIA had been completed.
Clinical and pathological findings
On April 28th, CFIA staff conducted an initial epidemiologi-
cal investigation on the herd and obtained samples for testing
at the National Centre for Foreign Animal Disease (NCFAD)
in Winnipeg. An assessment completed by the CFIA District
Veterinarian noted that approximately 25% of pigs in areas D
and E were exhibiting clinical signs of respiratory disease.
These signs were reported to have started around April 20th
and included a deep non-productive cough with an abdominal
effort and mild to moderate depression (Figure 2). A decrease
in feed consumption was also noted. The records revealed an
increase in percent mortality from 0.43% to 0.87% and 0.19%
to 2.04% in areas D and E, respectively through to the end of
April. Approximately 10% of pigs in areas D and E had been
treated parenterally by the owner with trimethoprim-sulfadoxine
(Trimidox; Vetoquinol Canada, Lavaltrie, Quebec). All other
production areas were clinically unaffected when the initial site
visit was completed by CFIA staff on April 28th. Under the
direction of the herd veterinarian, chlortetracycline medicated
premix (Aureomycin 110 G; Alpharma Canada, Mississauga,
Ontario) was added at a rate of 2.5 kg/tonne of complete feed
to the rations fed in areas D and E to alleviate concerns about
known secondary bacterial pathogens endemic to the herd.
On May 5th, the herd veterinarian conducted a full health
assessment. Ongoing daily communication with the producer
had revealed that the cough described earlier was now present
in other production areas. It was evident that the virus had
spread rapidly throughout the facility. Individual pigs within the
nursery-grow-finish (NGF) population within areas C through
G presented with a sudden onset clear oculo-nasal discharge,
sneezing, mild conjunctivitis, and a deep, dry, non-productive
cough with significant abdominal effort. Clinically affected pigs
in the NGF population were pyrexic, moderately depressed,
anorexic, and mildly dehydrated. In areas D and E morbidity
associated with ILI had declined significantly and was evident
in only 10% of animals. Mortality had returned to historical
levels less than 1%. In areas F and G only 5% of pigs presented
with ILI and mortality in these areas remained unchanged from
reports prior to the diagnosis of pandemic H1N1 2009 virus.
Approximately 10% of pigs in the nursery (area C) were
coughing, depressed, and dehydrated; the cough was mildly
productive suggesting a secondary bacterial infection. Mortality
had not increased in the nursery and remained consistent with
historical levels of well below 1%. As many as 10% of the oldest
piglets in the farrowing rooms (area B) had a mild cough and
were sneezing and only 1 sow showed any clinical signs of ILI.
In area A, 2 of the 24 gilts had a mild cough and were slightly
depressed; none of the sows showed any clinical signs of ILI.
Feed consumption by the sows (areas A and B) was unchanged
from previous visits.
Field necropsy examinations were performed on May 8th
during the limited cull of 475 grower/finisher pigs. Grossly,
pigs had poorly collapsed lungs with a rubbery texture and
mild interlobular edema in the dorso-caudal lobes. Multifocal
disseminated, dark red-purple, shrunken and firm individual
lobules, sharply demarcated from adjacent lobules and coalesc-
ing in cranio-ventral regions were also noted. A gross diagnosis
of lobular to coalescing broncho-pneumonia was made. There
was evidence of secondary bacterial infections with copious
purulent exudate in airways and severe consolidation with fibrin
and abscessation was observed in some pigs but was not a con-
sistent feature. Histopathologic examination of the trachea and
lungs revealed mild, chronic, non-specific tracheitis, moderate
broncho-interstitial pneumonia with perivascular and peribron-
chiolar lymphoid hyperplasia, mild multifocal necrotizing and
suppurative alveolitis, and subacute to chronic necrotizing to
hyperplastic bronchiolitis. Lesions were compatible with mild to
moderately severe infection with multiple respiratory pathogens
of PRDC including MH, PRRS, and secondary or opportunistic
bacterial pathogens. The bronchiolar lesions were characterized
by varying degrees of epithelial attenuation through to epithelial
regenerative hyperplasia with microabscessation and mild peri-
bronchiolar fibrosis, compatible with influenza infection in a
subacute to chronic reparative stage.
Figure 2. Eleven-week-old pre-grower pigs huddling with mild
depression and reluctance to rise (photo courtesy of Egan
Brockhoff).

CVJ / VOL 50 / NOVEMBE R 2009 1157
S P E C I A L R E PO R T
There was evidence both histologically and by PCR that this
herd dealt, on an ongoing basis, with M. hyopneumoniae and
PRRSv activity with relatively mild lesions of both. Based on the
herd history and these laboratory findings, the herd was fairly
stable with respect to the background ongoing components of
the PRDC, meaning that it did not typically experience dra-
matic mortality or morbidity attributable to these components.
This influenza virus infection resulted in increased morbidity
and mortality of short duration, with the acute infection passing
rapidly through the herd to the regenerative stage seen in the
pigs examined on May 8th.
During the week of May 11th, the herd veterinarian and
the CFIA veterinarians and staff reassessed the medical condi-
tion of the herd and collected additional samples. Nasal swabs
and blood samples were randomly collected from each of the
7 production areas for submission to NCFAD. The morbidity
and mortality rates observed at the end of April and during the
1st week of May had declined to the pre-influenza levels. Feed
consumption patterns within the herd were stable. The sow herd
continued to show stability and there was no evidence of ILI
in this group. The herd veterinarian and CFIA staff returned
to the barn the week of May 25th to perform repeat diagnostic
sampling and assess the health of the herd. Throughout areas A
to G the health of the pigs was unremarkable. The majority of
the pigs were bright, alert, and responsive. There was no cough
in the piglet population in area B and , 2% of the piglets
presented with a sneeze. There was no evidence of ILI in the
nursery population. The week of June 1st was the last time that
the herd was examined and diagnostic samples were procured.
Prior to the initiation of the depopulation process on June 4th,
the herd veterinarian detected no clinical evidence of ILI in any
of the production areas.
Results of diagnostic tests
Following the initial on-site investigation by CFIA staff, nasal
swab and serum samples from pigs in areas D, E, F, and G were
submitted to NCFAD. Twenty-four nasal swab specimens were
received on April 29th and RNA was immediately extracted for
testing using the Spackman single tube real-time polymerase
chain reaction (RT-PCR) assay targeting the M1 gene of influ-
enza A viruses that has been developed for avian influenza test-
ing (7). Preliminary results from this assay late in the evening
of April 29th showed that 3 of the 24 specimens produced
equivocal results and the remaining 21 were negative. As the sen-
sitivity of this assay for the novel swine-origin virus had already
been assessed as questionable by NCFAD, further testing using
conventional RT-PCR assays specific for the M gene (8) and the
H1 gene related to A/California/04/09 (National Microbiology
Laboratory, Public Health Agency of Canada, unpublished pro-
tocol) was simultaneously performed. Results available in the
morning of April 30th from the RT-PCR assays clearly identified
that 19 and 15 of the 24 nasal swabs specimens produced posi-
tive results for the M gene and H1 gene, respectively, confirming
infection with an H1 subtype influenza A virus.
A computer analysis of the M gene sequence for several of the
human swine-origin H1N1 isolates indicated that the routinely
used forward and reverse primers of the Spackman M1 gene
RT-PCR assay (7) were less than optimal and these were there-
fore re-designed. Using the modified primers, results available
in the early morning of May 2nd showed that 17 (71%) of the
original 24 swab specimens gave a positive and another 4 (17%)
samples gave suspicious results. This indicated that an influenza
A virus with an M gene segment similar to that identified in
several of the novel human isolates of pandemic H1N1 2009
was present (unpublished observations). The apparent prevalence
(9) of pandemic H1N1 2009 in areas D, E, F, and G based on
this sampling was estimated to be 87.5% [95% confidencial
interval (CI): 69.0–95.7]. The CFIA immediately notified the
World Organization for Animal Health (OIE) of these find-
ings. By May 4th NCFAD determined the partial sequence of
the M, H1, and N1 genes of the virus and confirmed that this
was the pandemic H1N1 2009 virus; these additional findings
were reported to the OIE on May 5th. Twenty-one of 24 nasal
swab samples yielded influenza A virus isolates and 1 isolate was
selected for full genome sequencing which was completed on
May 7th. Phylogenetic analysis determined that the genome was
99% homologous to the novel H1N1 influenza A virus caus-
ing illness in humans around the world. Sequence results were
submitted to GenBank on May 11th.
Of the 31 sera that were initially collected from barns D,
E, F, and G, 8 (25.8%) tested positive for antibodies to influ-
enza A virus. Five of these samples also had neutralizing anti-
body titers to the human isolate A/Mexico/InDRE4487/2009
(vH1N1 2009, kindly provided by the National Microbiology
Laboratory). The nasal swab samples from these serologically
positive animals gave either a negative or a weakly positive result
with the modified M gene RT-PCR. Collectively these results
suggest an initial infection 10 to 14 d earlier.
Follow-up testing using random sampling of the herd was
carried out 2, 4, and 5 wk after the initial date of sampling as
described under clinical findings. Although a proportion of nasal
swab specimens collected subsequent to April 28th were posi-
tive for influenza A virus nucleic acid, no live virus was isolated
from these samples. A decline in the proportion of samples
that produced positive or suspicious results by RT-PCR over
time was observed. Sampling completed during the week of
May 11th identified an apparent herd level prevalence of 13.6%
(95% CI: 9.6–18.8) with 29 out of 214 samples being posi-
tive or suspicious. Sampling completed the week of May 25th
identified an apparent herd level prevalence of 7.9% (95% CI:
5.2–11.9) with 20 of 252 samples positive or suspicious.
Sampling completed the week prior to depopulation (5 wk post
initial sampling) identified an apparent herd level prevalence
of 1.5% (95% CI: 0.5–4.4) with 3 of 198 samples being posi-
tive or suspicious. There was an increase in the proportion of
pigs seropositive to influenza A nucleoprotein over the same
time. On the date of initial sampling 8 of 31 (25.8%) samples
were seropositive, indicating previous exposure to influenza
A nucleoprotein in the sampled population. Two weeks later,
this proportion had increased to 54.4% (95% CI: 47.1–61.4)
and by 4 wk it had increased to 70.6 % (95% CI: 61.1–78.6).
A review of the serological results suggested that pigs in all
production areas were exposed to the virus within a relatively
short period of time. These time lines are consistent with a

1158 CVJ / VOL 50 / NOVEMBER 2009
RAPPORT SPÉCIAL
predicted incubation period of 1 to 3 d with rapid recovery in
4 to 7 d that is typical for classical swine influenza. These data
are presented in Figure 3.
Epidemiological investigation
Tracing of the movement of all pigs, pig products, objects
exposed to pig or pig products and humans associated with this
farm during the 21-day period prior to the onset of clinical signs
of respiratory disease observed was undertaken. The purpose
was to identify other swine farms or humans at risk of having
been exposed to pandemic H1N1 2009 virus and to attempt
to confirm the source of introduction of virus. Although our
initial hypothesis was that the contracted worker who returned
from Mexico was the most likely source of virus on this farm,
it was important to rule out the possibility of any other human
or swine source and confirm that no other farms were at risk
of being exposed.
The trace-out investigation did not identify any farms at risk
of exposure via the direct or indirect movement of humans or
animals. There was 1 shipment of 52 finished hogs to slaughter
on April 23 but these animals were shipped from production
areas F and G which were clinically unaffected on this date.
Ante- and postmortem examinations on these animals at the
slaughter plant were unremarkable. The most recent purchase
of animals was breeding gilts in February of 2009 from a pri-
vate purebred breeder. Prior to the delivery of gilts, and in the
subsequent months, there had been no ILI in the source herd.
The trace-in investigation did not identify any potential source
farms, ruling out the possibility that an unidentified swine
operation had been the source of virus.
It was confirmed that the individual hired to work on the
ventilation system in areas D and E had experienced ILI while
in the barn on April 14. This individual had returned from
Mexico on April 12 prior to international awareness of this
emerging disease. Retrospective investigation confirmed that
this virus had been circulating in Mexico for at least several
weeks prior to his return. Alberta Health Services (AHS) was
contacted to investigate the human illness associated with this
farm. The investigation of the hired individual, the farm fam-
ily, and other community members revealed that several cases
of pandemic H1N1 2009 were identified in the community in
April and May. A number of community members had recently
returned from travel to Mexico. Testing of the hired individual
by RT-PCR using a nasopharyngeal swab was negative for
influenza A using the probes, primers, and methods provided
by the Centers for Disease Control and Prevention (CDC)
Laboratory in Atlanta, Georgia. Of note, these nasopharyngeal
swabs were collected well after onset of clinical signs of ILI, at
days 11 and 19. Additional serological testing completed on
June 26, 2009 using microneutralization assay performed at
PHAC’s National Microbiology Laboratory (NML) and at the
CDC laboratory met PHAC’s definition of a confirmed case of
pandemic influenza virus (H1N1) 2009. There was also serologi-
cal evidence of previous infection with seasonal H1N1 influenza
virus. Although the hired individual wore a dust mask at times
when in the barn, the protective measures were not consistent
and did not suffice to prevent exposure of the pigs to the virus.
There were other individuals who had direct or indirect con-
tact with the pigs and exhibited ILI prior to the first observed
clinical signs in the pigs; however, it has not been possible to
confirm or rule out the potential for these individuals to have
been sources of the virus. There were H1N1-positive individuals
with an established epidemiological link to this farm who had
onset of symptoms after the swine were ill and the possibility of
swine to human transmission cannot be excluded in this group.
However, it is also possible that infection occurred as a result
of person-to-person spread either via contact with the hired
worker or other infected individuals in the community. The
public health investigation concluded that the hired individual
and possibly other individuals in direct contact with this herd
introduced the virus to the swine.
Occupational health and safety
The zoonotic potential of swine influenza viruses is well-
recognized (1). During the field epidemiological investigations,
diagnostic sampling and humane destruction activities both
CFIA and ARD staff entered the barn on multiple occasions. It
was recognized that these staff could be at an increased risk for
exposure to pandemic H1N1 2009 virus. Based on the specific
situation, current scientific knowledge of the novel virus and
after consultation with the Public Health Agency of Canada
(PHAC), Health Canada’s Workplace Health and Public Safety
Program (WHPSP) provided advice for all employees associated
with the response. In consultation with WHPSP and the CFIA,
Alberta Health and Wellness (AHW) and AHS provided advice
to ARD staff.
Considering the novel nature of this infection in both human
and swine populations, full personal protective equipment
(PPE) was recommended by WHPSP on May 5th; this included
N95 respirators, gloves, eye protection with seals around the
eyes, boots, hair covers and coveralls or other body suit be used
by all workers entering the barn. Due to the large amount of
potentially contaminated fluids in the air, the possibility that the
PPE may be dislodged, the perceived severity of human illness
associated with this infection at the time, and the precedent
set for anti-viral use during previous avian influenza responses,
antiviral medications were recommended for prophylaxis for
the duration of exposure plus 10 d. Prior to CFIA and ARD
staff completing sampling and depopulation activities, seasonal
influenza vaccine was recommended and antiviral medication
Figure 3. Estimated herd level prevalence of active viral
shedding (as detected by PCR) and seroprevalence
(to Influenza A) by week of investigation.
Week of Investigation
Estimated herd level
prevalence of viral
shedding by RT-PCR
testing
Estimated herd level
seroprevalence to
influenza A
Prevalence (%)
100
80
60
40
20
0
1 3 5 6

CVJ / VOL 50 / NOVEMBE R 2009 1159
S P E C I A L R E PO R T
(oseltamivir) offered to all staff in contact with the swine, but
not all workers took the medication. Canadian Food Inspection
Agency employees involved with the on-farm response were
assessed by an Occupational Health Medical Officer from
WHPSP for seasonal influenza vaccine, antiviral medication
as well as Tetanus/Diphtheria vaccine. Antiviral medication
was dispensed and the vaccines were administered by the
Occupational Health Nurse from WHPSP. Ongoing follow-up
was carried out to monitor adverse reactions to antiviral medi-
cation. Alberta Health Services provided similar occupational
health and safety support to ARD staff. If workers experienced
ILI after exposure to the quarantined premises they were advised
to isolate themselves until 24 h after symptoms had resolved
as a precautionary measure. As a nasopharyngeal swab is the
most sensitive diagnostic sample to confirm influenza infection,
exposed staff experiencing ILI were encouraged to contact the
AHS Medical Officer of Health to arrange testing.
In order to assess the potential human exposure to influenza
virus on this premises for both ARD and CFIA staff responding
to the outbreak, a health surveillance questionnaire was admin-
istered by the public health division of AHS. The telephone
questionnaire requested information on symptoms, onset and
duration, antiviral and vaccine use, protective equipment, and
breaches of the recommended protective practice. This public
health investigation was undertaken to determine whether ILI
was present in humans after being on the quarantined premises
and to assess the effectiveness of personal protective equipment
used to protect workers from exposure. It is recognized that
pig-to-human and human-to-pig transmission of influenza
virus occurs. It was also important to describe the type and
duration of exposure that could be associated with infection.
Response rates were excellent, with 93% of CFIA staff and 76%
of ARD staff completing the questionnaire.
Two confirmed cases of pandemic H1N1 2009 occurred in
workers who entered the barn on April 28th. These workers
became symptomatic within the expected incubation period
following exposure to the infected swine, and an investigation
into their illness supported a common source of exposure. These
individuals did not report any contact with a symptomatic
human prior to developing ILI. The investigation revealed that
there was an opportunity for transmission of the virus from the
infected swine to the workers. Although it is possible that the
exposed workers became infected from a human source that
has not been identified, there is epidemiological support for
zoonotic transmission in this situation.
The CFIA has determined that proper use of fitted full-
face respirators with P100/chemical combination cartridges
are indicated when working with influenza-infected animals.
This is due to: the physical exertion and positioning required
to carry out the required procedures, the heavy dust load of
the swine or poultry housing environment, the need for eye
protection (without fogging), and the common occurrence
of ammonia and other manure gases. Subsequent to these
identified cases, CFIA workers wore such protection. The
CFIA workers involved in high-risk sampling activities who
stated they followed the CFIA recommended procedures for
working on influenza contaminated premises (10) did not
become infected.
Anti-virals have been used in staff responding to previous out-
breaks of avian influenza. Anti-virals were recommended in this
outbreak for workers at risk of exposure due to the large amount
of potentially contaminated fluids in the air and the possibility
that PPE could be dislodged due to the working conditions.
None of the workers became symptomatic when taking anti-viral
medication at the time of exposure and in the 10 d following
exposure. However, it is difficult to attribute this difference to
the use of anti-virals in consideration of the small sample size.
In addition, the infected workers were potentially exposed to
higher concentrations of virus due to the higher prevalence of
sick pigs in the confined area on April 28th compared with later
workers. The importance of anti-virals for staff working with
animals infected with influenza viruses should continue to be
discussed, studied, and evaluated. Given that exposed staff are
known and can be followed closely, early treatment on develop-
ment of symptoms may be the best approach, as it minimizes
potential side effects from possibly unnecessary medication and
the promotion of resistance to anti-viral medication.
Approach to future cases of human pandemic
H1N1 2009 virus infection in swine herds in
Canada
The initial risk management decision by the CFIA and ARD
to place this herd under federal movement restrictions under
the Health of Animals Act was precautionary during a period
of significant public and global concern and scientific uncer-
tainty. At the time virus was confirmed on this farm, there was
a lack of information on the virulence of this virus in human
and pig populations. It was deemed prudent to conduct a full
epidemiological investigation and restrict movement until such
time as additional information was available and the risk to both
the swine and human populations of North America could be
assessed. As of August 7th other than Canada, only Argentina
and Australia have identified and reported infection of swine
with this virus.
Internationally, veterinary authorities are discussing the most
appropriate approach to manage influenza infections in swine
herds. Animal and public health authorities agree that influenza
virus is not a food-borne zoonosis and does not affect the safety
of properly cooked pork. Scientific evidence supports that live,
infective virus is not present beyond the respiratory tract, and
is most likely to be found in nasal and pharyngeal secretions
during the febrile period of illness, of 1 to 3 d post-exposure (3);
therefore, there is no risk of acquiring the virus from meat of
recovered animals. As with any raw meat, pork should always
be properly handled and cooked to eliminate a range of food
safety concerns. Acutely ill pigs that are shedding virus could
present a potential occupational risk to individuals handling
live animals, but the obvious clinical manifestations of illness in
affected animals (such as, respiratory signs, inactivity, decreased
feed intake) should preclude their shipment to slaughter until
they have recovered.
The OIE has stated that this virus is currently behaving in
the same fashion as other swine influenza A viruses and does

1160 CVJ / VOL 50 / NOVEMBER 2009
RAPPORT SPÉCIAL
not require restrictive trade or disease control policies to be
implemented. The Food and Agriculture Organization (FAO)
of the United Nations approaches the management of this
disease from a similar perspective. Public health authorities in
Canada, in line with World Health Organization (WHO) rec-
ommendations, have indicated that no extraordinary response
measures are needed or warranted in the human population to
control the spread of the virus at this time. Public health instead
provides advice about minimizing transmission, personal respi-
ratory hygiene and the role of prescription anti-virals, where
appropriate.
The number of humans who would have had contact with
an infected swine herd is extremely limited when considering
the opportunity for human-to-human transmission within any
given infected community. Consequently, the imposition of
strict control measures on swine herds while employing a more
measured approach in people may create the impression that
infected swine are more of a risk than infected people. This
is clearly inconsistent with the observations to date. Based on
current information, and the approach undertaken by public
health authorities for humans, the CFIA has modified its initial
approach of imposing federal quarantine restrictions on a swine
herd infected with, or exposed to, the pandemic H1N1 2009
virus. Under this new policy, a federal quarantine will not be
imposed unless there is evidence of a change in pathogenicity
of the virus in humans or pigs. The CFIA will assist with the
diagnostic characterization of any H1 influenza A virus isolated
by a non-CFIA laboratory and offer advice and assistance to the
provincial animal health authority if an infected herd is identi-
fied. The CFIA will lead a Federal/Provincial/Territorial con-
sultation on the most appropriate manner to manage infected
herds in collaboration with public health authorities, veterinary
practitioners, and industry stakeholders. Under the authority in
the Health of Animals Act and Regulations, the CFIA has the
legislative mandate and capacity to implement stringent con-
trol measures should a change in the virus increase the risk to
animal or public health and such measures become warranted.
The CFIA is also working in collaboration with the veterinary
authorities of the United States and Mexico on a protocol under
which the 3 countries will agree on the notification framework
and measures to be applied in a way to prevent trade restric-
tions. This approach is meant to minimize the economic impact
of regulatory movement restrictions on swine producers while
ensuring appropriate control mechanisms are in place.
Summary
This disease occurrence has highlighted the importance of ongo-
ing collaboration between animal and public health authorities
at all levels to ensure a timely and coordinated response to
emerging zoonotic diseases. Given the continued spread of this
virus in the human population, it is reasonable to predict that
additional cases of pandemic H1N1 2009 in swine herds in
Canada will be identified. Further dissemination of this virus
in pig populations may pose an additional risk for transmission
to humans in direct contact with clinically ill pigs. The need
for appropriate PPE for workers investigating and sampling on
suspect and confirmed infected farms is emphasized. Due to the
potential for human-swine influenza virus reassortment and the
development of a more virulent strain, the use of effective PPE
needs to be addressed in these circumstances. In addition, it
is important to emphasize that humans with ILI should avoid
contact with any influenza susceptible animal species.
Although anti-viral prophylaxis was utilized in this specific situ-
ation due to global uncertainty surrounding this emerging virus
at the time, early treatment of exposed individuals experiencing
ILI may be sufficient and more closely parallels the response in
human health care workers. Additional discussion and research
are required to determine whether workers in a barn environment
are more at risk of exposure and subsequent infection with this
virus than those in a health care setting. From a human health
perspective, the direct and indirect routes of human-to-human
transmission will continue to account for the vast majority of new
human infections. From an animal health perspective, evidence
from this outbreak and findings from experimental studies sug-
gest that in its current form this virus is unlikely to cause more
significant clinical disease in pigs than commonly observed with
classical SIV’s in Canada which is essentially a self-limiting infec-
tion confined to the respiratory tract with limited morbidity and
eventual recovery.
Acknowledgments
The authors thank Drs. Brian Evans, Francine Lord, Jim
Clark, Connie Argue, and Keith Campbell from the CFIA and
Dr. Gerald Hauer from ARD for their review of the manuscript,
and Drs. Kathleen Hooper-McGrevy, Carissa Embury-Hyatt,
and James Neufeld for their diagnostic work at NCFAD. The
authors acknowledge the diagnostic work completed at PHAC’s
National Microbiology Laboratory and at the CDC Laboratory
in Atlanta, Georgia. Diagnostic testing and advice provided by
the Alberta Provincial Laboratory of Public Health was essen-
tial to the human investigation. Field epidemiologist Rachel
McCormick contributed to the questionnaire design, data col-
lection, and analysis with regard to exposed workers. The CFIA
staff from Alberta North and other involved districts as well as
ARD staff are acknowledged for their hard work and attention
to detail during the investigation.
References
1. Olsen CW, Carey S, Hinshaw L, Karasin Al. Virologic and serologic
surveillance for human, swine and avian influenza virus infections
among pigs in the north-central United States. Arch Virol 2000;145:
1399–1419.
2. Vincent AL, Ma W, Lager KM, Janke BH, Richt JA. Swine influenza
viruses a North American Perspective. In: Maramorosch K, Shatkin AJ,
Murphy FA, eds. Advances in Virus Research. Vol. 72. Burlington: Acad
Pr, 2008:127–154.
3. Olsen CW. Swine Influenza. In: Straw BE, Zimmerman JJ, D’Allaire S,
Taylor DJ, eds. Diseases of Swine. 9th ed. Ames, Iowa: Blackwell Publ,
2006:469–482.
4. Myers KP, Olsen CW, Gray GC. Cases of swine influenza in humans:
A review of the literature. Clin Infect Dis 2007;44:1084–1088.
5. Gray GC, McCarthy T, Capuano AW, Setterquist SF, Olsen CW,
Alavanja MC. Swine workers and swine influenza virus infections.
Emerg Infect Dis 2007;13:1871–1878.
6. Van Reeth K. Avian and swine influenza viruses: Our current under-
standing of the zoonotic risk. Vet Res 2007;38:243–260.
7. Spackman E, Senne DA, Myers TJ, et al. Development of a real-time
reverse transcriptase PCR assay for type A influenza virus and avian H5
and H7 hemagglutinin subtypes. J Clin Microbiol 2002;40:3256–3260.

CVJ / VOL 50 / NOVEMBE R 2009 1161
S P E C I A L R E PO R T
8. Foucher RAM, Bestebroer TM, Herfst S, Van der Kemp L,
Rimmelzwaan GF, Osterhaus ADME. Detection of influenza A viruses
from different species by PCR amplification of conserved sequences in
the matrix gene. J Clin Microbiol 2000;38:4096–4101.
9. Brown LD, Cai TT, DasGupta A. Confidence limits for apparent preva-
lence using the Wilson binomial approximation: Interval Estimation for
a proportion. Stat Sci 2001;16:101–133.
10. Canadian Food Inspection Agency, Field Biocontainment Working
Group. Biocontainment and Safety Procedures for Working on Premises
Contaminated with Notifiable Avian Influenza Viruses, RDIMS
# 1305923, November 2007.
Practice Tips Conseils pour la clinique
Ken Keeler, DVM
Delton Veterinary Hospital
Edmonton, Alberta
Most clients don’t like needles,
and do not like to see their pets hurt.
Clients are often grateful if we can vaccinate their pet in a quick
and relatively painless manner. As much as possible, fill your
vaccine syringes before the client and patient come into the
examination room, and have the syringes in a discrete location
on the counter. Watching you load the syringes only heightens
the client’s anxiety. When it’s time to vaccinate the pet, and if
you don’t have an assistant in the examination room to help you,
ask the owner to hold the animal and also vigorously scratch
the pet behind its ears (the harder the better). If the owner is
very nervous of needles, they can close their eyes. Then quickly
vaccinate the pet. Don’t discard the empty syringes until the
examination is over, because sometimes the owner doesn’t think
you actually gave their pet its shots.
Most clients are really impressed by a timely
call back from the veterinarian
(even my own doctor does not call me).
Try to phone the owner later in the afternoon when they’re
more likely to be home, if things are not going well, book a
re-examination.
We receive more thank you letters from clients
after a dignified, caring euthanasia,
than we do after a successful medical
or surgical procedure.
In addition to your normal euthanasia routine, try to physically
touch the owner after the pet is deceased. A gently squeeze of
their hand, elbow, or shoulder, a handshake or even a hug (what-
ever seems appropriate) demonstrates your genuine empathy for
what the owner is experiencing. Then give the owner time alone
with their deceased pet. Most owners leave the room within
5 minutes. Write a short personal note on the sympathy card
and include that they “made the right decision” or “did the right
thing,” if the decision to euthanize seemed a difficult one — this
owner will likely return to you when they get a new pet.
Say anal glands, not anal sacs.
Sacs sometimes sounds like sex.
Here are 2 ways to help differentiate between
back pain and abdominal pain/cramps in a dog.
When palpating the dog for pain, only touch the region you
are testing. For example, if you’re checking for back pain, pal-
pate the back but do not steady the dog by placing your other
hand under the abdomen! When you’ve finished your physical
examination, place the dog on the floor at the far end of the
examination room. Dogs without back pain will usually give a
full body shake after walking over to the owner, somewhat like
they shake off water after a bath. A dog with back pain will only
shake the head and shoulders, stop, and then shake its pelvis
and tail.