Preventive Veterinary Medicine 123 (2016) 161167

Contents lists available at ScienceDirect

Preventive Veterinary Medicine

journal homepage: www.elsevier.com/locate/prevetmed

Epidemiological factors associated to spread of porcine epidemic

diarrhea in Japan
Yosuke Sasaki a,b, , Julio Alvarez c , Satoshi Sekiguchi b,d , Masuo Sueyoshi b,d ,
Satoshi Otake e , Andres Perez c
a

Organization for Promotion of Tenure Track, University of Miyazaki, Miyazaki, Japan

Center for Animal Disease Control, University of Miyazaki, Miyazaki, Japan
c
Department of Veterinary Population Medicine, College of Veterinary Medicine, University of Minnesota, MN, USA
d
Department of Veterinary Sciences, Faculty of Agriculture, University of Miyazaki, Miyazaki, Japan
e
Swine Extension & Consulting, Inc., Niigata, Japan
b

a r t i c l e

i n f o

Article history:
Received 16 May 2015
Received in revised form 29 October 2015
Accepted 1 November 2015
Keywords:
Biosecurity
Porcine epidemic diarrhea
Risk factor
Swine
Transmission
Japan

a b s t r a c t
Porcine epidemic diarrhea (PED) is an emerging disease of pigs that has recently led to large numbers
of piglet deaths in a number of countries of Eastern Asia and The Americas. The objective of the present
study was to identify and compare risk factors associated with PED infection in locally and non-locally
PED-exposed farms in Japan. A questionnaire was administered to a convenience selection of pig farms
located throughout Japan. Questionnaires were administered between November 2013 (when the rst
case was reported in Japan) and August 2014. PED-positive farms (cases, n = 124) were asked to provide
information on their status (positive or negative) and select herd management practices for the two weeks
prior to onset of PED clinical signs. Negative farms (controls, n = 128) were given the same questionnaire
and asked herd management practices for the two weeks prior to a given reference date. This date was
assigned based on the date of PED occurrence in the town/prefecture in which the farm was located.
Case and control farms were categorized as locally exposed if they were located within a 5 km radius
from a PED-infected farm and non-locally exposed, otherwise. Logistic regression analysis was used to
identify factors associated with PED infection. Two separate regressions were done for locally exposed
and non-locally exposed farms using PED status (positive/negative) as the dependent variable. PED in
locally-exposed farms was associated (P < 0.05) with increased farm size (in 100 pig increments), shorter
distances to the closest PED-positive farm (less than1,001 m), and a disinfectant contact time of less
than 20 min. In non-locally exposed farms, PED was associated (P < 0.05) with increased feed truck visits
to the farm, no visit of the veterinarian, and again a disinfectant contact time of less than 20 min. These
ndings suggest that the mechanisms of PED spread in Japan were different for farms closer to case-farms
compared to farms that were further away from PED cases. These results will contribute to understanding
the epidemiology of the disease in Japan and will ultimately aid in designing and implementing effective
prevention and control strategies in Japan and other regions epidemically infected by the PED virus.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Porcine epidemic diarrhea virus (PEDv) is an enveloped, singlestranded, positive-sense RNA virus of the genus Alphacoronavirinae
in the family Coronaviridae that is related to the transmissible gastroenteritis (TGE) virus (Hofmann and Wyler, 1989; Saif et al.,
2012). The PEDv emerged as a global threat to the swine industry in

Corresponding author at: 1-1 Gakuen Kibanadai-nishi, Miyazaki 889-2192,

Japan. Fax: +81 985 58 7864.
E-mail address: yskssk@cc.miyazaki-u.ac.jp (Y. Sasaki).
http://dx.doi.org/10.1016/j.prevetmed.2015.11.002
0167-5877/ 2015 Elsevier B.V. All rights reserved.

2013, when a number of epidemics were reported in many important swine-producing countries of North America and Eastern Asia
that were previously believed to be PEDv-free. These countries
included US, Canada, and Mexico, including Korea, Taiwan, and
Japan (Cima, 2013; More, 2013; Stevenson et al., 2013; Chen et al.,
2014; Park et al., 2014; Hanke et al., 2015; Song et al., 2015). PEDv
infection results in high morbidity and high mortality in <10 days
old piglets, and decreased growth performance for those that survive infection (Huang et al., 2013; Mole, 2013; Stevenson et al.,
2013; Alvarez et al., 2015b). PED was rst reported in Japan in
the 1990s, and a PED live vaccine was approved in 1996 (Sueyoshi
et al., 1995). Since then, only isolated and relatively unimportant

162

Y. Sasaki et al. / Preventive Veterinary Medicine 123 (2016) 161167

outbreaks have been recorded. However, in October 2013 the rst

PED outbreak was reported in Japan (Okinawa prefecture) after an
interval of seven years. The virus rapidly spread across the country
with 817 PED cases conrmed across 38 prefectures as of August
31, 2014 (Ministry of Agriculture, Forestry and Fisheries of Japan,
2014b). The PED strain detected in 2013 was genetically different
from those circulating in the 1990s in Japan, and similar to the virus
found in the US that same year (Nguyen et al., Submitted). Despite
substantial human and nancial resources being allocated to elucidate the mechanisms associated with the introduction and rapid
spread of the virus (Ministry of Agriculture, Forestry and Fisheries of
Japan, 2014b), there is still limited knowledge on the epidemiology
of PEDv infection in Japan.
Variables previously identied as risk factors for PED spread
in Japan include biosecurity practices, pig movement, and management during pig loading. These variables are also considered
risk factors for PED spread worldwide (Zimmerman et al., 2012;
Lowe et al., 2014). Feed has also been identied as a potential
source for PEDv-spread in Japan following the infection with PED
of a number of nucleus farms (Sasaki, personal communication).
In these farms biosecurity protocols were particularly rigorous so
feed was thought to be the only mechanism of invasion. The danger of feed transmission is consistent with studies performed in
the US in which feed materials such as spray-dried porcine plasma
were suspected to cause PED spread (Gerber et al., 2014; Pasick
et al., 2014). Given the high infectiousness and low infectious dose
of PEDv, proximity to an infected farm may also increase the risk
of infection due to local spread (through aerosol transmission or
contaminated fomites). This route is suggested by ndings in Japan
and the U.S. (Alvarez et al., 2015a; Sasaki et al., 2015).
Here, we hypothesized that factors associated with high risk of
infection in Japan were different for farms in areas in which other
infected farms were present (referred to as locally exposed farms)
compared to farms that were further away from infected farms
(referred to as non-locally exposed farms). Results will help to
quantify the dynamics of PED spread and, ultimately, support the
design and implementation of PED prevention and control measures in Japan and other epidemically-infected countries.

2. Materials and methods

2.1. Data collection
The study population included all pig farms that are consulted by
clinical veterinarians afliated with the Japan Association of Swine
Practitioners (JASP). The JASP includes 52 veterinarians working in
approximately 500 farms, which are approximately 10% of all herds
in Japan. The country had 5270 herds in February 2014 (Ministry
of Agriculture, Forestry and Fisheries of Japan, 2014a). Each veterinarian serves a range of 121 farms. In December 2014, all
veterinarians in the JASP were invited to complete a questionnaire
in relation to all PED positive farms (case farms) in their practice
and an approximately equal number of control farms. Case farms
were swine farms with a clinical presentation of PED and laboratory
conrmation (RT-PCR). Control farms were farms from the same
practice (selected by the veterinarian) with no history of PED infection. A questionnaire was administered to a convenience selection
of pig farms located throughout Japan.
The questionnaire was designed to collect data regarding herd
management practices. The questionnaire (Table 1) contained ten
sections detailing herd management practices in the following
areas: farm information, pig movement, use of own/borrowed
equipment, feed, site visitors, manure application, biosecurity practices, pests and wildlife, biosecurity at pig loading/unloading, and
disease status. The questionnaire (available upon request to the

rst author) was pre-tested by a selected number of veterinarians

regarding content, interpretation of questions, and responses.
Questionnaires were lled out between November 2013 (after
the rst case of the epidemic was detected in Japan) and August
2014. PED-positive farms (cases) were asked to provide information
on their herd management practices for the two weeks prior to the
initial detection of PED clinical signs in the farm. Negative farms
(controls) were asked to provide the information for the two weeks
prior to a reference date. This date was set for each control farm
based on the rst occurrence of PED in the town and prefecture in
which the farm was located. Specically, if PED had been detected
in the same town, the reference date was dened as the date in
which the rst PED-positive farm was identied in that town. If
PED had not been detected in the town of location of the control
farm, the reference date was set as the date of the rst outbreak
in the closest town and, when no disease had been reported in the
closest town, in the prefecture. If no PED had been reported in the
same prefecture, the reference date was set as the date of detection
of PED in the closest PED-positive prefecture. In Japan, all farms in
which symptoms compatible with PED are detected must report
this occurrence to the regional livestock hygiene center. If fecal
samples from those farms showed RT-PCR positive (or PEDv RNA
was detected by RT-PCR in the fecal samples), those were conrmed
as a case farm. PEDv infection in all study farms was conrmed by
RT-PCR and immunohistochemistry. We converted the farm name
to an index number.
Case and control farms were categorized as locally-exposed
(located within a 5 km radius from a PED-infected farm) and nonlocally exposed, otherwise. The 5 km radius was selected because it
was the spatial distance that maximized the observed-to-expected
ratio of cases in Japan and the U.S., as reported elsewhere (Alvarez
et al., 2015a; Sasaki et al., 2015).
2.2. Statistical analysis
All information was entered into a generic worksheet program
and recoded into categorical data (nominal or ordinal level) for further analysis. Data were analyzed using the SAS software ver. 9.4
(SAS Institute Inc., Cary, NC, USA).
Two logistic regression analysis models were tted to identify
factors associated with PED: one in locally exposed and one in
non-locally exposed farms. For both models, the response variable
was PED-status (positive, negative). A univariate logistic regression
analysis was performed to assess the association between PED status and each independent variable in the questionnaire (Table 1).
Variables associated (P < 0.25) with PED were subsequently tested
together in a multivariable logistic regression model. A backwards
elimination procedure, that used a P < 0.05 selection threshold for
the removal of non-signicant variables, was used to identify main
effects for the nal model. Once main effects were obtained, twoway interactions were tested on the basis of biological plausibility.
Although absence of independence associated with time and space
may have been accounted for by the study design, location (four
regional blocks in Japan) and time (month in which the farm provided information) were alternatively included as random effects
in the multivariable model using the GLIMMIX procedure. The
strength of association between variables and the response was
quantied by computing the odds ratios (OR) and their corresponding 95% condence intervals (CI).
3. Results
3.1. Descriptive statistics
Data were received from 42 veterinary practices, providing
information for 258 (129 positive, 129 negative) farms. Some (2%,

Y. Sasaki et al. / Preventive Veterinary Medicine 123 (2016) 161167

163

Table 1
Variables and denition used for the analysis.
Item
Farm information
Total number of pigs
Production system
Distance from closest PED-positive farm

Category and denition

Total number of pigs in the farm (Continuous variable)
Farrow-to-nish operation, Farrow-to-wean operation, Wean-to-nish operation, Nucleus operation
Locally-exposed farms: 0100 m, 101500 m, 5011000 m, 10012000 m, 20015000 m
Non-locally-exposed farms: 500110,000 m, 10,00120,000 m, 20,000 m or more

Pig movement
Pig addition
Pig removal
Use of own/borrowed equipment

Yes/No (Farms added/did not add any type of pigs into the farm)
Yes/No (Farms removed/did not remove any type of pigs into the farm)
Yes/No (Farms borrowed/did not borrow equipment from another swine farm)

Feed
Visit of feed truck
Articial milk

Number of feed truck visited to the farm (Continuous variable)

Yes/No (Articial milk included/did not include spray-dried porcine plasma)

Site visitors
Rendering truck
Veterinarian
Other visitor
Manure application
Biosecurity practices
For personnel and employees

For business visitors

Pests and wildlife

Biosecurity at pig loading/unloading

Wash and disinfection

Loading place
Contact time
Disease status
Disease status

Vaccine program

Yes/No (Rendering truck visited/did not visit to the farm)

Yes/No (Veterinarian visited/did not visit to the farm)
Yes/No (Other visitor such as electric, gas and diesel company visited/did not visit to the farm)
Applied on communal manure dropoff spot, Applied on land belonging to the farm,
Sold to crop farmer, Miscellaneous
High biosecurity (Applied more than 5 items described below),
Intermediate biosecurity (Applied 3 to 4 items),
Low biosecurity (Applied less than 2 items)
The questionnaire items were: Required to shower before operation, Change clothes and footwear before operation,
Wear rubber boots, Change rubber boots in each barn, Required to shower after operation, Regularly wash and disinfect
clothes/boots
High biosecurity (Applied more than 6 items described below),
Intermediate biosecurity (Applied 4 to 5 items),
Low biosecurity (Applied less than 3 items)
The questionnaire items were: Required to shower before operation, Change clothes and footwear before operation,
Wear rubber boots, Change rubber boots in each barn, Required to shower after operation, Regularly wash and disinfect
clothes/boots, Must wait 24 h or more after visiting another site with pigs
High (Applied more than 5 items described below),
Intermediate (Applied 3 to 4 items),
Low (Applied less than 2 items)
The questionnaire items were: Use rat and mouse poison bait stations, Use traps for rodent control, Use y control for
buildings, Pig/hog buildings are bird proof, Onsite feed storage area prevents access by wildlife, Never seen bird within
barn, Never seen rodent within barn, Never seen cat within barn
High (Applied more than 7 items described below),
Intermediate (Applied 3 to 6 items),
Low (Applied less than 2 items)
The questionnaire items were: Washing applied to the chute oor after each loading/unloading, Disinfection applied to
the chute oor after each loading/unloading, The driver never step in an area at any time during or after the
loading/unloading process, Wash truck before loading, Disinfect truck before loading, Wash and disinfection performed
separately before loading, Wash trucks and trailers prior to visiting a site after leaving another swine site, Wash animal
area of truck, Disinfect animal area of truck, Wash outside area of truck, Disinfect outside area of truck
Yes/No (Loading/unloading area for pigs is/is not located more than 30 m from swine housing)
Yes/No (Allow/Not allow disinfectant a contact time of more than 20 min)
High (Had diseases more than 4 items described below),
Intermediate (Had diseases 1 to 3 items),
Low (No diseases)
The questionnaire items were: Actinobacillus pleuropneumonia (APP), Porcine reproductive and respiratory syndrome
(PRRS), Mycoplasma, Salmonella, Swine dysentery, Transmissible gastroenteritis (TGE), Porcine circovirus associated
disease (PCVAD), Streptococcus suis, Atrophic rhinitis (AR), Erysipelas, Glassers disease, Toxoplasma, Staphylococcus,
Aujeszkys disease (AD), E. coli, Clostridium, Roundworm, Mange, Swine u, Porcine proliferative enteritis (PPE)
High (Vaccinated more than 9 items described below),
Intermediate (Vaccinated 6 to 8 items),
Low (Vaccinated less than 5 items)
The questionnaire items were: APP, PRRS, Mycoplasma, Salmonella, Swine dysentery, TGE, PCVAD, Streptococcus suis, AR,
Erysipelas, Glassers disease, Toxoplasma, Staphylococcus, AD, E. coli, Clostridium, Roundworm, Mange, Swine u, PPE

n = 6) farms were excluded due to an incomplete questionnaire

(n = 3) or a known movement-related cause of PED infection (introduction of weaned pigs from a PED positive farm, n = 3). Therefore,
data from 252 (124 positive, 128 negative) farms were used for
the analyses here. Numbers of positive and negative farms were
85 (62.0%) and 52 (37.0%) in locally exposed farms, and 39 (33.9%)
and 76 (66.1%) in non-locally exposed farms, respectively (Table 2).

Therefore, 68.5% of all positive farms were assumed to have been

locally exposed farms, whereas the remaining 31.5% positive
farms were included in the non-locally exposed group farms. In
both categories, farrow-to-nish operation was the main production type (78.1% (107 out of 137) for locally exposed farms
and 73.9% (85 out of 115) for non-locally exposed farms, respectively).

164

Y. Sasaki et al. / Preventive Veterinary Medicine 123 (2016) 161167

Table 2
Number of PED-positive and -negative farms in locally exposed and non-locally exposed farms stratied by production type and proximity to PED-positive farms.
Locally exposed farms

Non-locally exposed farms

Positive

Negative

Positive

Negative

107
21
9

62.6%
57.1%
66.6%

37.4%
42.9%
33.3%

85
20
7
3

35.3%
25.0%
57.1%
0%

64.7%
75.0%
42.9%
100.0%

Distance from closest PED-positive farm

32
0100 m
20
101500 m
22
5011000 m
10012000 m
18
20015000 m
45
500110,000 m

10,00120,000 m

20,000 m or more

84.4%
80.0%
68.2%
38.9%
44.4%

15.6%
20.0%
31.8%
61.1%
55.6%

25
25
65

44.0%
32.0%
30.8%

56.0%
68.0%
69.2%

Total number of farms

62.0%

38.0%

115

33.9%

66.1%

Production type
Farrow-to-nish
Farrow-to-wean
Wean-to-nish
Nucleus

137

3.2. Factors associated with PED in locally exposed farms

Only 8 of the 20 assessed factors were associated (P < 0.25) with
PED status in the univariate analysis (Table 3) and subsequently
tested in the multivariable logistic regression model. Three factors (total number of pigs, distance from closest PED-positive farm
and contact time for disinfectant) were retained in the nal model
(P < 0.05), and no signicant interaction was detected (Table 4).
Inclusion of space or time as random effects did not improve the
tness of the model as suggested by an increase in the AIC values
Signicant associations had the following effects: (1) an increase
of 100 pigs in farm size increased PED odds by 2.7% (OR: 1.027),
(2) farms located within 100 meters from an infected farm were
at an almost 12 times higher odds for PED than farms located at
least 1 km apart from the nearest infected farm (OR = 11.7), and (3)
farms that did not allow the disinfectant to have a contact time
longer than 20 min had an almost 3 times (OR: 2.748) higher odds
of being PED-infected than those that allowed for such contact time.
3.3. Factors associated with PED in non-locally exposed farms
After a backward elimination of the 10 factors associated with
PED in the univariate analysis (P < 0.25, Table 3), three factors (number of feed truck visits to the farm, visit of veterinarians, and contact
time for disinfectant) remained signicant (P < 0.05; Table 5). No
signicant interactions were found. Inclusion of space or time random effect did not improve the model as demonstrated by a higher
value of the AIC. According to those results, an increase in one feed
truck visit to the farm increased the odds for PED by 16% (OR = 1.16),
whereas having visits from a veterinarian the 2 weeks before the
outbreak or reference date reduced the odds of PED by nearly 70
percent (OR = 0.31). Odds of being PED-positive increased over two
and a half times for farms that did not allow the disinfectant a
contact time longer than 20 min (OR = 2.63).
4. Discussion
This is the rst study aimed at comparing factors associated with
PED infection for farms that were closely located to PED-infected
farms as opposed to further away. Our ndings are consistent with
the hypothesis that local transmission of PED was associated with
factors different from those associated with long distance spread.
Distance to the closest PED-infected farm was signicantly associated with PED outbreaks in the locally exposed farms. Odds of PED
almost doubled when the distance decreased from 501 to 1000 m
to 101500 or 101500 to 0100 m (Table 4). Although between-

farm aerosol transmission has never been documented, PEDv RNA

has been detected in air samples up to 10 miles from an infected
farm (Alonso et al., 2014). Other routes of transmission that may be
related with distance between neighboring farm include contaminated trucks (Lowe et al., 2014), equipment, or people. Distance
to closest PED-positive farm was not signicantly associated with
positivity to PED in non-locally exposed farms, suggesting that the
risk associated with proximity to an infected farm is no longer
signicant at distances larger than 5 km.
In locally exposed farms, farm capacity (or size), as indicated by
the total number of pigs in the farm, was considered a risk factor
for PED spread. Unmeasured characteristics (latent) within farm
capacity such as number of personnel on the farm or number of
visits of loading trucks are likely associated with farm size. Therefore the number of contacts with the farm, and the associated risk,
is expected to increase with the size of the farm. However, farm
size was not a risk factor for non-locally exposed farms, suggesting
that this factor may only be critical when the potential source of
infection is local. Consequently, large farms should increase their
biosecurity measures as they pertain to contacts in the event of a
PED outbreak in the area. Those measures may include, for example, the proper use of disinfectants, as indicated by the model
results. Allowing disinfectant a contact time of more than 20 min
may decrease the risk of PED introduction into both locally-exposed
farms and non-locally exposed farms. Shulaw and Bowman (2001)
reported that disinfectants must have sufcient contact time with
the surfaces to which they are applied in order to allow them to
kill the pathogens with which they are designed for. In non-locally
exposed farms, a visit of a veterinarian was considered a protective factor against PED occurrence. A possible explanation for this
nding may be that veterinarian visits may be associated with an
increase in the biosecurity measures in the farm, or that a more
proactive producer who spots something wrong may invite the
veterinarians to come quickly and implement additional measures
sooner.
Number of feed truck visits to the farm during the 2 weeks prior
to the outbreak was considered a risk factor for PED only in nonlocally exposed farms. Some have reported that transport vehicles
for swine and feed may contain PEDv (Dee et al., 2014a,b; Lowe
et al., 2014; OIE, 2014). In Japan, it has been found that PED positive
farms have shared feed trucks with previously PED-infected farms
(Ministry of Agriculture, Forestry and Fisheries of Japan, 2014b).
This suggests that trucks may have acted as the vehicle of infectious for the virus. In addition, either the feed or the feedbags being
transported may have also been contaminated (Dee et al., 2014a,b;
Bowman et al., 2015). It is believed that transport pallets that car-

Y. Sasaki et al. / Preventive Veterinary Medicine 123 (2016) 161167

165

Table 3
Bivariate association between PED in locally exposed and non-locally exposed farms and factors hypothesized to inuence that risk in Japan.
Locally exposed farms

Non-locally exposed farms

Item

P-value

P-value

Farm information
Total number of pigs
Production system
Distance from closest PED-positive farm

137
137
137

0.014
NS
0.002

115
115
115

0.160
NS
NS

Pig movement
Pig addition
Pig removal
Use of own/borrowed equipment

137
137
137

NS
NS
NS

115
115
115

0.037
NS
NS

137
117

0.170
0.070

115
104

0.001
0.047

137
137
137
137

0.222
NS
NS
0.218

115
115
115
115

NS
0.047
NS
NS

Biosecurity practices
For personnel and employees
For business visitors
Pests and wildlife

137
137
137

NS
NS
NS

115
115
115

0.041
0.226
NS

Biosecurity at pig loading/unloading

Wash and disinfection
Loading place
Contact time

137
137
137

NS
NS
0.057

115
115
115

0.148
NS
0.050

Disease status
Disease status
Vaccine program

137
137

0.125
NS

115
115

NS
0.094

Feed
Visit of feed truck
Articial milk
Site visitors
Rendering truck
Veterinarian
Other visitor
Manure application

Table 4
Multivariable association between PED in locally exposed farms and factors hypothesized to inuence that risk in Japan.
Item

PED positive%

P-value

Odds ratio (95%CI)

Total number of pigs, 100 pigs

Distance from closest PED-positive farm
0100 m
101500 m
5011000 m
10012000 m
20015000 m

137

0.002
0.001

1.027 (1.0191.036)

32
20
22
18
45

84.4
80.0
66.7
42.1
53.1

Contact time (Allow/Not allow disinfectant a contact time of more than 20 min)
42
50.0
Yes
95
67.4
No

11.687 (11.59011.785)
6.451 (1.66025.066)
3.470 (1.08111.139)
NS
Reference
0.003

Reference
2.748 (1.1076.828)

Table 5
Multivariable association between PED in non-locally exposed farms and factors hypothesized to inuence that risk in Japan.
Item

PED positive%

P-value

Oddsratio (95%CI)

Number of feed truck visited to the farm

115

0.001

1.159 (1.0651.262)

Visit of veterinarians
Yes
No

38
77

21.1
40.3

0.026

0.309 (0.1110.860)
Reference

22.7
40.8

0.043

Reference
2.632 (1.0426.645)

Contact time (Allow/Not allow disinfectant a contact time of more than 20 min)
44
Yes
71
No

ried feedbags may have not been sufciently disinfected at the

beginning of the PED outbreak in Japan (Ministry of Agriculture,
Forestry and Fisheries of Japan, 2014b).
Although we failed to identify an association between articial
milk with spray-dried porcine plasma and PED in the multivariable
analysis, the variable was associated with PED in the univariate
analysis. It should be considered that failure to identify the association may be simply because other factors were sufcient to explain

occurrence of PED (King et al., 1998). Indeed, PCR analysis of spraydried porcine plasma imported to Japan from March to May 2014
revealed that a large proportion of the reduced number of samples tested (7/8) were positive, although a bioassay on nursery
pigs was negative (Ministry of Agriculture, Forestry and Fisheries of
Japan, 2014b). Given the limited sensitivity of the bioassay, negative results cannot be considered as a proof of the lack of association
between feed and disease (Gerber et al., 2014; Opriessnig et al.,

166

Y. Sasaki et al. / Preventive Veterinary Medicine 123 (2016) 161167

2014; Pasick et al., 2014; Pujols and Segals, 2014; Bowman et al.,
2015). Although PCR-positive results demonstrate that feed was
contaminated with viral RNA, it is unclear whether the virus would
still be infectious.
Production system was not related to PED occurrence in both
locally-exposed and non-locally exposed farms suggesting that all
farm types were at similar risk of being PED-infected. However, it
may well be that some wean-to-nish operations may have failed
to identify PED-like signs because those signs may not be as evident
in older pigs. Earlier studies reported that there is an age-dependent
resistance to pathogenic PED virus infection in pigs (Shibata et al.,
2000). Clinically silent nishing pigs shipped into slaughter plants
were strongly suspected to transmit PED in Japan.
Proximity to infected neighboring farms in addition to other
management factors including biosecurity practices and herd size
have been already identied as important factors associated with
an increased risk of infection at the farm level in other important
swine diseases such as PRRS (Mortensen et al., 2002; Lambert et al.,
2012; Velasova et al., 2012). Our results, on the other hand, suggest
that the mechanisms for PED spread in Japan were different for
farms nearer to previously infected farms, compared to those that
were further, something difcult to assess in the case of a widely
distributed disease as PRRS. Results stress the need for changing
and adapting biosecurity practices consistently with the circumstances of PED spread. Our results must be interpreted with caution
given potential biases associated with our reference population
(JASP farms) compared with the whole Japanese swine population
and our sample size, that could limit our ability to detect significant risk factors. Still, we are not aware of any systematic bias
in JASP farms with regards to non-JASP farms and, to our knowledge, this is the rst time that risk factors associated with risk of
disease in farms locally exposed are compared with those of nonlocally exposed in the midst of a PED epidemic. These results could
contribute to understanding the epidemiology of the disease in the
country and, ultimately, to design and implement effective prevention and control strategies in Japan and other regions epidemically
infected by the PED virus.
Conict of interest
None of the authors of this paper has a nancial or personal
relationship with other people or organizations that could inappropriately inuence or bias the content of the paper.
Acknowledgments
The authors gratefully thank the cooperative producers for completing the questionnaires and providing their data for use in the
present study, and the veterinarians and staff at the Japan Association of Swine Practitioners for their assistance. This work was
supported by a KAKENHI(Grants-in-Aid for Scientic Research)
from the Japan Society for the Promotion of Science (No. 26870454).
References
Alonso, C., Goede, D.P., Morrison, R.B., Davies, P.R., Rovira, A., Marthaler, D.G.,
Torremorell, M., 2014. Evidence of infectivity of airborne porcine epidemic
diarrhea virus and detection of airborne viral RNA at long distances from
infected herds. Vet. Res. 45, 73.
Alvarez, J., Goede, D., Morrison, R., Perez, A., 2015a. Spatial and temporal
epidemiology of porcine epidemic diarrhea (PED) in the Midwest and
Southeast regions of the United States. Prev. Vet. Med. (in press).
Alvarez, J., Sarradell, J., Morrison, R., Perez, A., 2015b. Impact of porcine epidemic
diarrhea on performance of growing pigs. PLoS One 10, e0120532.
Bowman, A.S., Krogwold, R.A., Price, T., Davis, M., Moeller, S.J., 2015. Investigating
the introduction of porcine epidemic diarrhea virus into an Ohio swine
operation. BMC Vet. Res. 11, 38.
Chen, Q., Ganwu, L., Stasko, J., Thomas, J.T., Stensland, W.R., Pillatzki, A.E., Gauger,
P.C., Schwartz, K.J., Madson, D., Yoon, K.J., Stevenson, G.W., Burrough, E.R.,

Harmon, K.M., Main, R.G., Zhang, J., 2014. Isolation and characterization of
porcine epidemic diarrhea viruses associated with the disease 2013 outbreak
among swine in the United States. J. Clin. Microbiol. 52, 234243.
Cima, G., 2013. Fighting a deadly pig disease. J. Am. Vet. Med. Assoc. 243, 467470.
Dee, S., Clement, T., Schelkopf, A., Nerem, J., Knudsen, D., Christopher-Hennings, J.,
Nelso, E., 2014a. An evaluation of contaminated complete feed as a vehicle for
porcine epidemic diarrhea virus infection of nave pigs following consumption
via natural feeding behavior: proof of concept. BMC Vet. Res. 10, 176.
Dee, S., Neil, C., Clement, T., Christopher-Hennings, J., Nelso, E., 2014b. An
evaluation of a liquid antimicrobial (Sal CURB ) for reducing the risk of porcine
epidemic diarrhea virus infection of nave pigs during consumption of
contaminated feed. BMC Vet. Res. 10, 220.
Gerber, P.F., Xiao, C.-T., Chen, Q., Zhang, J., Halbur, P.G., Opriessnig, T., 2014. The
spray-drying process is sufcient to inactivate infectious porcine epidemic
diarrhea virus in plasma. Vet. Microbiol. 174, 8692.
Hanke, D., Jenckel, M., Petrov, A., Ritzmann, M., Stadler, J., Akimkin, V., Blome, S.,
Pohlmann, A., Schirrmeier, H., Beer, M., Hper, D., 2015. Comparison of porcine
epidemic diarrhea viruses from Germany and the United States. Emerg. Infect.
Dis. 21, 493496.
Hofmann, M., Wyler, R., 1989. Quantitation, biological and physicochemical
properties of cell culture-adapted porcine epidemic diarrhea coronavirus
(PEDV). Vet. Microbiol. 20, 131142.
Huang, Y.W., Dickerman, A.W., Pineyro, P., Li, L., Fang, L., Kiehne, R., Opriessnig, T.,
Meng, X.J., 2013. Origin, evolution, and genotyping of emergent porcine
epidemic diarrhea virus strains in the United States. MBio 4, 00737813.
King, V.L., Koketsu, Y., Reeves, D., Xue, J., Dial, G.D., 1998. Management factors
associated with swine breeding-herd productivity in the United States. Prev.
Vet. Med. 35, 255264.
Lambert, M.., Arsenault, J., Poljak, Z., DAllaire, S., 2012. Correlation among genetic,
Euclidean, temporal, and herd ownership distances of porcine reproductive
and respiratory syndrome virus strains in Quebec, Canada. BMC Vet. Res. 8, 76.
Lowe, J., Gauger, P., Harmon, K., Zhang, J., Connor, J., Yeske, P., Loula, T., Levis, I.,
Dufresne, L., Main, R., 2014. Role of transportation in spread of porcine
epidemic diarrhea virus infection, United States. Emerg. Infect. Dis. 20,
872874.
Ministry of Agriculture, Forestry and Fisheries of Japan, 2014a. Statistics of
Agriculture, Forestry and Fisheries. (accessed 24.08.15.).
Ministry of Agriculture, Forestry and Fisheries of Japan, 2014b. Porcine Epidemic
Diarrhea. (accessed 24.08.15.). .
Mole, B., 2013. Deadly pig virus slips through US borders. Nature 499, 388.
Mortensen, S., Stryhn, H., Sogaard, R., Borklund, A., Stark, D., Christensen, J.,
Willeberg, P., 2002. Risk factors for infection of sow herds with porcine
reproductive and respiratory (PRRS) virus. Prev. Vet. Med. 53, 83101.
OIE World Organization for Animal Health, 2014. Technical factsheet Infection
with porcine epidemic diarrhoea virus. http://www.oie.int/
leadmin/Home/eng/Our scientic expertise/docs/pdf/A factsheet PEDV.pdf.
Opriessnig, T., Xiao, C.T., Gerber, P.F., Zhang, J., Halbur, P.G., 2014. Porcine epidemic
diarrhea virus RNA present in commercial spray-dried porcine plasma is not
infectious to naive pigs. PLoS One 9, e104766.
Park, S., Kim, S., Song, D., Park, B., 2014. Novel porcine epidemic diarrhea virus
variant with large genomic deletion, South Korea. Emerg. Infect. Dis. 20,
20892093.
Pasick, J., Berhane, Y., Ojkic, D., Maxie, G., Embury-Hyatt, C., Swekla, K., Handel, K.,
Fairles, J., Alexandersen, S., 2014. Investigation into the role of potentially
contaminated feed as a source of the rst-detected outbreaks of porcine
epidemic diarrhea in Canada. Transbound. Emerg. Dis. 61, 397410.
Pujols, J., Segals, J., 2014. Survivability of porcine epidemic diarrhea virus (PEDV)
in bovine plasma submitted to spray drying processing and held at different
time by temperature storage conditions. Vet. Microbiol. 174, 427432.
Saif, L.J., Pensaert, M.B., Sestak, K., Yeo, S.-G., Jung, K., 2012. Coronaviruses. In:
Zimmerman, J.J., Karriker, L.A., Ramirez, A., Schwartz, K.J., Stevenson, G.W.
(Eds.), Diseases of Swine. John Wiley & Sons, West Sussex, UK, pp. 501524.
Sasaki, Y., Alvarez, J., Sekiguchi, S., Sueyoshi, M., Otake, S., Perez, A., 2015. The
spatial dynamics of porcine epidemic diarrhea (PED) spread in Miyazaki
prefecture, Japan. In: The 14th International Society for Veterinary
Epidemiology and Economics, November 37, 2015, Merida, Mexico.
Shibata, I., Tsuda, T., Mori, M., Ono, M., Sueyoshi, M., Uruno, K., 2000. Isolation of
porcine epidemic diarrhea virus in porcine cell cultures and experimental
infection of pigs of different ages. Vet. Microbiol. 72, 173182.
Shulaw, W.P., Bowman, G.L., 2001. Disinfection in On-Farm Biosecurity Procedures.
http://www.gov.on.ca/OMAFRA/english/livestock/biosecurity/facts/
procedures.htm.
Song, D., Huang, D., Peng, Q., Huang, T., Chen, Y., Zhang, T., Nie, X., He, H., Wang, P.,
Liu, Q., Tang, Y., 2015. Molecular characterization and phylogenetic analysis of
porcine epidemic diarrhea viruses associated with outbreaks of severe
diarrhea in piglets in Jiangxi, China. 2013. PLoS One 10, e0120310.
Stevenson, G.W., Hoang, H., Schwartz, K.J., Burrough, E.R., Sun, D., Madson, D.,
Cooper, V.L., Pillatzki, A., Gauger, P., Schmitt, B.J., Koster, L.G., Killian, M.L.,
Yoon, K.J., 2013. Emergence of porcine epidemic diarrhea virus in the United
States: clinical signs, lesions, and viral genomic sequences. J. Vet. Diagn. Invest.
25, 649654.
Sueyoshi, M., Tsuda, T., Yamazaki, K., Yoshida, K., Nakazawa, M., Sato, K., Minami,
T., Iwashita, K., Watanabe, M., Suzuki, Y., 1995. An immunohistochemical
investigation of porcine epidemic diarrhoea. J. Comp. Pathol. 113, 5967.

Y. Sasaki et al. / Preventive Veterinary Medicine 123 (2016) 161167

Velasova, M., Alarcon, P., Williamson, S., Wieland, B., 2012. Risk factors for porcine
reproductive and respiratory syndrome virus infection and resulting
challenges for effective disease surveillance. BMC Vet. Res. 8, 184.
Zimmerman, J.J., Beneld, D.A., Dee, S.A., Murtaugh, M.P., Stadejek, T., Stevenson,
G.W., Torremorell, M., 2012. Porcine reproductive and respiratory syndrome

167

virus (Porcine arterivirus). In: Zimmerman, J.J., Karriker, L.A., Ramirez, A.,
Schwartz, K.J., Stevenson, G.W. (Eds.), Diseases of Swine, 31, 10th ed. Blackwell
Publishing, pp. 461468 (chapter 31).