key: cord-0747777-s6yit1wj
authors: nan
title: Collection and review of updated scientific epidemiological data on porcine epidemic diarrhoea
date: 2016-02-17
journal: EFSA J
DOI: 10.2903/j.efsa.2016.4375
sha: 4020dedb5c1616fb4e9cb159581e76f7aa5ee3d5
doc_id: 747777
cord_uid: s6yit1wj

Porcine epidemic diarrhoea (PED) is a non‐zoonotic viral disease of pigs caused by a coronavirus and characterised by watery diarrhoea and weight loss. PED is not notifiable to the EU or World Organisation for Animal Health listed but it is notifiable at the national level in Finland, France, Ireland and Sweden. PED case reports from seven countries and PED surveillance and monitoring activities in thirteen countries were reported. This information was combined with an extensive literature review to provide an update on global PED occurrence, circulating strains and impact in 2014–2015. PED confirmed cases have been reported in North America, South America, Asia and Europe. PED virus (PEDV) sequences originating from EU pig herds indicate that the strains currently in circulation share nearly 100% sequence identity and have greater than 99% sequence identity with the reference INDEL (insertion/deletion) strain USA/OH851/2014. In 2014–2015, greater genetic variability has been reported in strains circulating in Asia compared with EU Member States and a non‐INDEL strain has been detected in the Ukraine in 2014. Data on impact confirms that mortality is higher in suckling piglets and diarrhoea is observed in all age groups. The reported impact is in agreement with that reported in EFSA AHAW Panel (2014) indicating that the impact of recently reported PED outbreaks in Asia and the USA seems to be more severe than that described in EU countries, although the impact of different PEDV strains is difficult to compare between one country and another, as impact is dependent not only on pathogenicity but also on factors such as biosecurity, herd size, farm management, sanitary status or herd immune status.

1.

In recent years, many porcine epidemic diarrhoea (PED) outbreaks caused by different virus strains have been reported by several countries worldwide with various degrees of impacts on pig production. Only a few Member States of the European Union (EU) have shared information or reported PED clinical cases and/or PEDV seropositive animals, the overall impact of PED in the Union being very limited.

The recent EFSA Scientific Opinion on PED and emerging porcine deltacoronavirus reports that the Czech Republic, Italy, Hungary, Germany and the United Kingdom have reported PED recently with mainly quite mild clinical cases and/or PED virus (PEDV) being isolated or PED seropositive animals being found. In the past weeks information has been made available about sporadic occurrence of PED infected animals in the Netherlands and possibly in other Member States.

PED is not a notifiable disease in the EU and it is not among the World Organisation for Animal Health-listed diseases. There is no evidence that the disease is causing significant health or production problems in the European pig farming system. The current epidemiological situation requires risk managers to obtain up-to-date consolidated information on the distribution and consequences of PED in the EU and worldwide.

Therefore, in the context of Article 31 of Regulation (EC) No 178/2002, the European Commission would like to ask EFSA to provide scientific assistance to the Commission following these terms of reference:

1) Guidance on PED data to be collected in Member States in order to optimise the coordination necessary to address the requests below. This may include a basic harmonisation of the case definition, the eligible diagnostic methods, the desired data sets and the frequency of reporting, as well as guidance on epidemiological investigations to facilitate data collection and to carry out the relevant epidemiological analysis. 2) An analysis of the epidemiological data and metadata available in the Member States and in recent scientific literature within and outside the EU, focusing on the occurrence of infection with different PEDV strains/types, as well as on the actual morbidity and mortality rates and severity of clinical disease so as to quantify the direct impact on pig production.

In addition, the outcome of the analysis of the above data should allow EFSA to predict possible epidemiological trends of the evolution of the disease within and outside the EU.

Additional information 1.2.1. Pig production in Europe Eurostat (2014) reports that the pig-producing countries of Europe can be divided into groups based on the type of production and level of commercialisation.

Large-scale fattener herds, characterised by having more than 400 pigs and no sows, are mainly found in Belgium, Denmark, Germany, Spain, Italy, Luxembourg, the Netherlands, Finland, Sweden and the United Kingdom. These 10 countries represent three quarters of EU pork meat production and 69% of the EU pig herd. Fattener herds with more than 400 pigs represent around 1.1% of the pig farms in Europe.

Small-scale/backyard pig farms, characterised by less than 10 pigs, are mainly found in Romania, Hungary, Croatia, Slovenia, Lithuania, Latvia, and Bulgaria. In these countries, pigs are often raised for personal consumption. Small/backyard farms account for 73% of the pig farms in Europe.

Large-scale breeder herds, characterised by more than 400 pigs and 100 sows, are also found in the Czech Republic, Estonia, Ireland, Greece, Cyprus and Portugal. These countries represent half of the EU sow herd and 0.6% of the pig farms in Europe.

In other countries, for example in Poland, there is a transition from backyard pig farming to larger commercial operations.

In Figures 1 and 2 the difference in farm structures between countries can be clearly seen when the density of pig holdings is compared with the density of piglet heads. When interpreting the results presented in this document, it is important to consider the diversity of pig production and the economic importance of pig production in certain countries. Updated epidemiological data on PED www.efsa.europa.eu/efsajournal four web conferences, data-reporting guidelines (Annex A) and procedures were discussed and agreed. The reporting guidelines describe two data models. The first data model (Herd Level) allowed the reporting of pig herds which meet the PED case definition (see Section 2.2.1). This data model was used to collect temporal and spatial information, a description of the production system, reported mortality and morbidity and availability of sequencing data. The second data model (Reporting of serological surveys and other testing) allowed the reporting of monitoring and surveillance activities for PED in European Economic Area (EEA) countries. The data were submitted using Microsoft ® Excel templates via the EFSA Data Collection Framework (DCF 1 ). Austria, Belgium, Spain, France, Italy and the Netherlands reported Herd Level data, while Austria, Belgium, Denmark, Finland, France, Ireland, the Netherlands, Norway, Sweden and the United Kingdom reported data on PED testing activities. Germany provided information only on the number of herds meeting the case definition not through the DCF, but by e-mail. No data were submitted from Bulgaria, Croatia, Cyprus, Czech Republic, Greece, Hungary, Latvia, Lithuania, Luxembourg, Malta, Poland, Portugal, Romania, Slovakia and Slovenia.

In response to Term of Reference (ToR) 1, guidance on PED data to be collected was developed in consultation with experts of the EFSA Network on PED. The data-reporting guidelines which describe the two data models and the terminology definitions can be found in Annex A. In order for a herd to be reported for the Herd Level data collection, it had to meet the PED case definition. Data were provided for herds affected between 1 January 2013 and 30 September 2015. The second data model could be used to report the results of serological surveys and other PEDV testing not covered by the case definition (for example, negative results or testing of healthy pigs for monitoring or trade purposes). Results could be reported for any time period of testing and for testing of previously stored sera. Outbreaks already included in the Scientific Opinion of the EFSA Panel on Animal Health and Welfare, on PED and emerging porcine deltacoronavirus (EFSA AHAW Panel, 2014) did not need to be reported.

Confirmed cases: Following suspicion based on clinical signs (described in Annex A), a confirmation of viral infection is necessary, by any of the following tests: RT-PCR, antigen enzyme-linked immunosorbent assay (ELISA), immunohistochemistry, electron microscopy, immunoelectron microscopy or virus isolation.

Herd case definition: Any herd with one or more confirmed cases.

Morbidity, mortality and diagnostic testing information was requested separately for each of the following age categories, suckling piglets (newborn to weaning), weaned pigs (from weaning up to 60-80 days of age), fatteners (70-90 days of age, approximately 30-40 kg to slaughter) and adults (breeding pigs, including both sows and boars).

An extensive literature review was undertaken in order to collate information on PED for the time period between October 2014 and October 2015. More recent publications identified by the EFSA Network on PED have been included as indicated in footnotes to the respective literature review summary tables. The primary purpose of the extensive literature review was to obtain data in order to update tables 2, 7 and 8 of the 'Scientific Opinion on porcine epidemic diarrhoea and emerging porcine deltacoronavirus' of the EFSA Panel on Animal Health and Welfare (AHAW), published in 2014 (EFSA AHAW Panel, 2014) .

The protocol for the extensive literature review is described in Appendix C. The review was designed to address three main questions: Question 1: Where have cases/outbreaks of pigs with laboratory confirmation of PED been reported in 2014-2015?

Question 2: Sequence identity of full length and partial sequence data of PED isolates from pigs worldwide in 2014-2015 and their similarity to INDEL strain USA/OH851/2014, INDEL strain GER/ L00719/2014 and/or CV777 or other strains (e.g. USA/Colorado/2013) Question 3: What is the impact on pig production of PED reported worldwide for pigs with laboratory confirmation of PEDmortality, morbidity and duration of disease in the four age classes (suckling, weaned, fatteners, adults).

Fourteen bibliographic databases were queried in order to identify information on PED that has been published within the specified time period, appropriate to address these three questions. Additionally, the CVOs of the EU Member States were asked to provide any relevant information on country reports on PED that may have been published within the last few months. In addition a search was performed on websites of Veterinary Authorities, other governmental organisations and International Organisations and disease reporting databases as well as a search using Google, in order to identify web sites which contained .pdf files reporting on PED. Information to be retrieved was limited to publications in English, German, French, Italian and Spanish.

The information extracted from the reviewed literature can be found in Appendix A.

In response to ToR 2, countries reported data via the DCF. These data and the results of the literature review are summarised in the sections below.

Occurrence of PED in 2014-2015

The literature review identified 19 reports of PEDV detection published between October 2014 and October 2015, while two additional reports published after October 2015 have been also included in the respective table, in order to include information that was made available after the literature review. PEDV-confirmed cases/outbreaks have been reported in North America (USA, Canada), South America (Colombia), Asia (Japan, China, South Korea, Thailand, Taiwan, Vietnam) and Europe (Portugal, Spain, Slovenia, Germany, France, Belgium, Italy, Ukraine) (Appendix A).

PED is only notifiable in the following Member States: Finland, France, Ireland and Sweden (and is likely to become notifiable in the United Kingdom from late 2015/early 2016). The EFSA Network on PED representing EEA countries voluntarily reported 245 case reports of pig herds meeting the PED case definition and 71 pig herds with PCR confirmation of PEDV-genome were reported via e-mail by a private laboratory. The PEDV-confirmed cases were reported by Austria, Belgium, Spain, France, Italy, the Netherlands and Germany. Laboratory testing for PED is often not centralised within each MS and there exists a diverse range of laboratories testing for PED both in the private and public sector; there is also no official EU reference laboratory for this disease.

Belgium, Spain, Italy, Germany and the Netherlands are among the largest producers of pork in Europe, while globally, China, USA, Brazil, Germany and Vietnam are among the highest producers of pork (number of heads in average 2012-2013). 2 Farms in these countries are generally large commercial operations under continuous veterinary supervision.

Based on the data reported by the EFSA Network on PED via the DCF, Figure 3 shows the PED case reports by date of sampling and type of production system, prior to 30 September 2015. For the reported herds, the first confirmed sample was taken in March 2014. A greater number of case reports were reported in the beginning of 2015 and this could be due to an increasing number of clinical cases, increased vigilance for signs or changes in laboratory protocols and the range of tests applied to samples. In May 2015, the number of PED case reports decreased until August 2015, while a slight increase was again observed in September 2015. As the disease in not notifiable to the EU (only at national level in a few countries), any temporal distribution should only be considered to be indicative as there are likely to be infected herds which have not been reported or tested. Forty-three percent of case reports were finisher herds and 31% were grower-producer herds; all pig herds were raised indoors with only 3% having outdoor access. As clinical signs are more noticeable in piglets, it is surprising that the majority of herds for which confirmed cases were reported are finishers; however, this production system is the dominant type of pig production in the reporting countries; moreover, frequent movements of animals both on and off farm in rearing herds may increase the risk of PEDV introduction.

Based on the low number of case reports that have been submitted for the 2014-2015 period, the number of PED-affected herds in the EU is very low.

Virus strains circulating in 2014-2015

The literature review retrieved information on full length PEDV genome sequences from virus strains from pigs in Belgium, France, Germany, Ukraine, USA, Mexico, China, Japan, South Korea, Taiwan, Thailand and partial sequences (predominantly the S gene), from strains sampled in China, Korea, Portugal, Vietnam and the USA (Appendix A). In addition, the EFSA Network on PED provided information on full-length sequences of strains from France and Spain and partial sequences of strains obtained in Austria, Belgium, Spain, Italy and the Netherlands. Vlasova et al. (2014) clinical signs and no piglets had died (Wang et al., 2014) . This PEDV variant is characterised by three deletions and one insertion in the S gene. Phylogenetic analysis of full-length genome sequences from EU MS indicated that INDEL strain USA/ OH851/2014 shared greater than 99% sequence identity with the German, French, Belgian and Spanish strains (Hanke et al., 2015; Grasland et al., 2015; Stadler et al., 2015; Theuns et al., 2015) (Table 1) .

The German INDEL strain GER/L00719/2014 isolated from a fattening pig farm in May 2014 (Hanke et al., 2015) had 99.9% identity with strain no. KR011756 from France (Grasland et al., 2015) . The Belgian strain (BEL/15V010/2015) had 99.9% identity with L00721 also isolated in Germany in 2014 and 97% identity with the older European strain CV777 (Theuns et al., 2015) .

In Italy, comparative analysis of partial sequences of the RdRp and M genes and the total spike (S1) gene of the circulating strains responsible for the outbreaks in 2014 showed a high nt identity with the S-INDEL strain USA/OH851/2014 from the USA (98. 7, 99.8, 99.3-99.5%, respectively) and with the strains detected in Germany (100, 100, 99.7%) (Boniotti et al., 2016) . This differs from the PEDV strains previously circulating from 2007 to 2012 in Italy, which showed a high genetic variability.

In the Netherlands, partial sequence analysis (orf1b + S segment, 1,400 bp) of 11 PCR products from 11 farms showed that all viruses were > 99.5% similar to the German isolate and the US INDEL USA/OH851/2014 strain.

In Austria, sequencing of the complete or nearly complete S gene (> 4 kb) obtained from affected farms showed over 99.5% sequence identity to other recent PEDV strains from Western and Central Europe (Steinrigl et al., 2015) .

In Portugal, analysis of amplified sequences of the S gene shared a very high (99.0%) identity with the INDEL strain USA/OH851/2014 and were identical (100%) to the strains recently reported in Germany (Mesquita et al., 2015) .

Greater genetic diversity is seen in Asian strains compared to European MS strains ( Table 1) . Analysis of 38 full genome sequences from Japan found 4 strains, which clustered with North American INDEL strain in Clade II and that 34 clustered with Korean and North American strains in Clade I (Suzuki et al., 2015) . The authors concluded that recent PED outbreaks in Japan were caused by PEDV that has a common origin with those circulating in other swine-producing countries. Also in Japan, a strain isolated from the Tottori Prefecture had highest sequence identity to Iowa 103 and clustered with the North American isolates in Clade I (Masuda et al., 2015; Murakami et al., 2015) . The authors speculated that the large deletion in the S gene may affect virulence. A full-length genomic sequence from an isolate obtained from pigs with diarrhoea in South Korea in March 2014 has 99.7% sequence Strains isolated from 5 regions in China between 2011 and 2014 were sequenced and the 11 whole genomes exhibited nucleotide sequence identities ranging from 97.5% to 99.7%, with no insertions or deletions (Sun et al., 2015) . These strains clustered with strains from China and North America which had been isolated after 2008. These genomes of PEDV field isolates from China showed a wide range of genetic variation and highlight the risk of further PEDV outbreaks caused by novel variants.

In the USA, a strain isolated in Minnesota in January 2014 had nucleotide identities of 99.9% to USA/Colorado/2013 strain and clustered within Clade II (Marthaler et al., 2014) . The clinical presentation was reported as equal to or more severe than the presentation of cases caused by the prototype PEDV Colorado/2013.

Closer to the EU, a strain (Ukraine/Poltava01/2014) isolated from an outbreak in a farm in Poltrava in the Ukraine in 2014 had 99.8% sequence identity to strains USA/Kansas29/2013 and USA/Colorado30/2013 and only 98.5% sequence identity with the German strains all of which cluster within Clade II. This Ukraine strain is, therefore, distinct from the strains currently circulating in the MS.

Descriptive epidemiology for the herds where a full length genome sequence or published partial sequences are available indicated that the impact is variable between farms (Table 2) even though the sequence identity is high between these PEDV strains. From the literature review there is limited information available about the production systems, breeds and ages of the animals, clinical signs and impact and conclusions of the outbreak investigation. The linking of epidemiology and outbreak data with sequence data is required to understand the contribution of insertion/deletion events to PEDV virulence and impact and predict epidemiological trends.

The data model allowed the reporting of the proportion of animals with clinical signs and the number of dead animals for the four age classes (suckling piglets, weaned pigs, fatteners and adult animals) as well as the duration of clinical signs on the affected holding (in days). Box-and-whisker plots of the percentage of symptomatic animals and of PED-associated mortality by age category are shown in Figures 4 and 5 . The descriptive statistics for these values are presented in Table 3 . It is important to note the number of missing values; for all variables this is high and very high for the mortality figures. For various reasons, follow-up visits to, or contact with, confirmed holdings to collect information on the severity or otherwise of the disease may not occur. It is for this reason that impact analysis is difficult for diseases which are not notifiable and this is particularly problematic if the impact is low. Additionally, impact analysis would need to consider baseline values, in the absence of the disease, accounting for production types and age classes, and this information is often missing. High proportions of missing data (in particular for mortality) may be an indicator for disease with low production impact. Due to missing data, the results should be viewed with caution; however, based on the reported data, there is some indication that mortality was higher in the suckling piglet age group with a mean of 18%, and the range of mortality in this age group from 0% to 84% was greater than in other age groups. The mean proportion of pigs with clinical signs was between 48% and 60% across the four age groups. The impact indicators reported (morbidity, mortality, duration of clinical signs) varied between farms. The duration of clinical signs on the farm for affected herds was between 6 and 90 days with a mean of 23 days.

In addition to the data submitted by the EFSA Network on PED; information on the impact of PED on pig production was retrieved from the literature review and is summarised in Table A .3 in Appendix A. This table includes information from France, Germany, Spain, Canada, the USA and Colombia. Usually, morbidity and mortality were higher in piglets than in adult animals, in some cases reaching 100%. The duration of clinical signs of the disease was very variable, ranging from 2 days in individual adult animals to up to 15 weeks in some piglet groups.

Documenting production losses usually requires the analysis of high quality data that have been collected in combination with suitable baseline production values. A few studies have attempted to quantify PED impact in production terms. This information can be found in Table A .3 in Appendix A. Effects reported include reduced feed consumption (and, respectively, milk output in affected post-partum sows), reduced average daily weight gain and increased feed conversion ratio, a longer fattening period, reduced number of piglets per litter and reproductive losses (abortions, returns to heat, etc.). Two studies reported that the period for performance indicators to return to the values observed before the outbreaks were 6 and 20 weeks, respectively (Dastjerdi et al., 2015; Stadler et al., 2015) .

The reported impact is in agreement with EFSA AHAW Panel (2014) which indicates that the impact of recently reported PED outbreaks in Asia and the USA seems to be more severe than that described in EU countries, but the impact of different PEDV strains is difficult to compare between one country and another, as it is dependent not only on pathogenicity but also on factors such as biosecurity, herd size, farm management, sanitary status or herd immune status. 

Concerning possible suspected sources of PEDV infection for affected farms and specific disease control methods applied in relation to PED, the information reported for confirmed herds is categorised in Table 4 . Among the six countries that reported relevant information in Table 4 , five indicated transportation as a suspected source of PEDV infection. Concerning disease control measures, management, biosecurity and/or cleaning and disinfection measures were reported in all countries that reported herds with confirmed PED (through the DCFfor Germany this information is not available, as it reported only the number of confirmed cases by e-mail). From the literature review it seems that biosecurity plays an important role in prevention and that particular attention should be paid also to the cleaning of transportation trucks. The bottom and top edges of the box indicate the intraquartile range, the marker inside the box indicates the mean value and the line inside the box indicates the median value, the whiskers that extend from each box indicate the range of values that are outside of the intraquartile range, points that are a distance of more than 1.5 9 intraquartile range from the box are considered to be outliers. Figure 5 : Box-and-whisker plot of porcine epidemic diarrhoea-associated mortality by age category Updated epidemiological data on PED 3.5.

Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Ireland, Italy, the Netherlands, Norway, Sweden and the United Kingdom reported surveillance and monitoring activities. Among these countries, Denmark, Finland, Ireland, Norway and the United Kingdom did not confirm any PED cases in their country's pigs through these activities between June 2013 and 30 September 2015.

Activities based around serological testing for PED are listed in Tables 5 and 6. Table 6 describes surveys reporting the results as prevalence estimates. Sweden has had an active surveillance programme for PED between 1993 and 2005, but is not now performing surveillance for PED. Denmark had active surveillance using an in-house ELISA to detect PEDV-specific antibodies which tested approximately 2,500 swine sera each year between 2000 and 2006 with no positive results reported during this period of time. In the Netherlands, in 2014, a retrospective serological survey of 838 serum samples from sows obtained from herds and slaughterhouses resulted in two positive samples as determined by an indirect ELISA and subsequent confirmation using a virus neutralisation test. In Belgium, in 2014, sow serum samples from 12 herds randomly selected from each province were tested using an immunoperoxidase monolayer assay (IPMA) method and no samples tested positive.

The United Kingdom reported the results of a seroprevalence survey conducted in 2013 using serum samples obtained at the slaughterhouse ( Table 6 ). The results showed that a small proportion (9%) of the pigs sampled at the slaughterhouse were seropositive to PEDV using a blocking ELISA (Cheney et al., 2014) . The blocking ELISA test used for this survey (van Nieuwstadt and Zetstra, 1991) is known to be capable of detecting antibody to virulent US and historic Great Britain PEDV strains; however, the authors identified some uncertainties with respect to the antibody results and falsepositive results are recognised to occur. As indicated in Table 7 , there have been no PED PCR-positive samples from diarrhoeic UK pigs tested in passive surveillance activities from 2013 to 2015.

AFBI Stormont Veterinary laboratory tested 200 randomly selected sera obtained from pigs in Northern Ireland mainly at slaughter during late 2014 and 2015, for the purpose of evaluating a commercial antibody ELISA kit. No samples tested positive.

France conducted a prevalence survey in 2014 of serum samples from 300 breeding sows in 30 herds using an indirect ELISA test (Table 6 ). The test performance parameters were estimated from pig farms samples from Canada. After correcting for herd clustering and test sensitivity and specificity the frequency of seropositive sows was 3.6% with a 95% confidence interval of 1.55-6.47%.

Austria, Belgium, Denmark, Estonia, Finland, France, Ireland, Italy and the United Kingdom reported activities related to laboratory testing (of potentially infected materials, e.g. intestine, faeces, rectal swabs) where PED was suspected on farms with clinical signs (Table 7 ). In Belgium, in 2015, faecal samples and intestinal content from suspect carcasses were tested by RT-PCR and one positive result was obtained; ELISA analysis of serum from some of the herds also tested positive (Table 7 and  Table 5 , respectively). France is performing a passive surveillance on herds presenting clinical signs (surveillance programme implemented as the disease is notifiable as an emergent disease since July 2014). In Germany, there are no official surveillance or monitoring activities for PED but private laboratories are providing testing upon request.

There is uncertainty about the sensitivity and specificity of existing antibody tests when used for the assessment of exposure to emerging PEDV strains. Many tests were developed using strains circulating in Europe in the 1970s and 1980s (for example, CV777 isolated from a swine breeding farm in Belgium in 1977 (Pensaert and de Bouck, 1978) ) and the ability to detect the new strains from outside Europe requires further investigation.

As a consequence, there has been a ring-test performed between five EU national veterinary public health institutes, organised by DTU National Veterinary Institute, Denmark (partners within CoVetLab; CVI, Central Veterinary Institute, the Netherlands; Animal & Plant Health Agency (APHA),UK; ANSES, Agence nationale de s ecurit e sanitaire, de l'alimentation, de l'environnement et du travail, France; SVA, National Veterinary Institute, Sweden). An interlaboratory comparison of assays for detection of antibodies against PEDV has been conducted using a single panel of sera (> 50 samples) collected from the field both in Europe and in the USA and also from animal experiments including infections by PEDVs from both Europe and the USA. A variety of assays, based on either IPMA or ELISA technologies, including a commercial test kit, were included in the ring-test.

Differences were found in the ability of the various assays to detect anti-PEDV antibodies and also the specificity of the assays seemed to be variable. An 'in-house' blocking ELISA used at DTUVet, Denmark was used as reference and seemed to have the overall highest sensitivity and specificity. In Updated epidemiological data on PED www.efsa.europa.eu/efsajournal this 'in-house' blocking ELISA, a cell culture-grown isolate of PEDV (BR1/87) was used as antigen in combination with a pig polyclonal anti-PEDV antiserum. More than 2,000 pig sera from Denmark (which has not experienced cases of PED) tested negative in this assay, but pigs experimentally infected with PEDV (both EU and US strains) were shown to seroconvert within 10 days post-infection. Other tests, including a commercial test kit commonly used in many European laboratories were found to have lower specificity and/or sensitivity, giving rise to false positive or negative results.

There is uncertainty about serological test performance; therefore, negative or positive results without confirmation of the presence of the virus and/or possibly existing relevant clinical or epidemiological information should be interpreted with caution.

In collaboration with experts nominated by the CVOs of the EU Member States and the EFSA Scientific Network for Risk Assessment in Animal Health and Welfare, data collection and reporting guidelines for two data models were developed, one at herd level, used to report data for an epidemiological analysis and the second used to collect data to describe surveillance and monitoring activities. During this process, a harmonised case definition was agreed for PED-confirmed herds based on clinical signs and laboratory confirmation of the PEDV.

The data collected by the EFSA Network on PED were combined with data retrieved from an extensive literature review covering the period of October 2014-October 2015 and collated information on PEDV occurrence, circulating strains and impact.

PED-confirmed cases/outbreaks have been reported in North America, South America, Asia and Europe in 2014-2015. Herds meeting the case definition for PED were reported voluntarily by Austria, Belgium, Spain, France, Italy, the Netherlands and Germany. Thirteen countries reported PED surveillance and monitoring activities and among these, Denmark, Finland, Ireland, Norway and the United Kingdom did not confirm any PED cases in their country's pigs through these activities between June 2013 and 30 September 2015. No data were submitted from Bulgaria, Croatia, Cyprus, Czech Republic, Greece, Hungary, Latvia, Lithuania, Luxembourg, Malta, Poland, Portugal, Romania, Slovakia and Slovenia.

Sequencing of virus isolates from EU pig herds indicated that the strains currently in circulation shared nearly 100% sequence identity and had greater than 99% sequence identity with the reference INDEL strain USA/OH851/2014. Italy has reported that the observed high sequence identity contrasts with PEDV strains previously circulating from 2007 to 2012 in Italy, which showed a high genetic variability. However greater genetic variability is observed in Asia compared with EU MS and a non INDEL strain has been detected in an outbreak in Ukraine in 2014. This highlights the possibility of further outbreaks in pig populations naive to a novel PEDV variant.

Data provided by the EFSA Network on PED and information from the literature review confirmed that mortality is higher in suckling piglets and diarrhoea is observed in all age groups for strains circulating globally in 2014-2015. Two published studies reported the time period required for the return of production to pre-outbreak levels, and this was variable for the described outbreaks (6 and 20 weeks, respectively) (Stadler et al., 2015 , Dastjerdi et al., 2015 .

These findings are in agreement with those reported in EFSA AHAW Panel, (2014) that the impact of recently reported PED outbreaks in Asia and the USA seems to be more severe than that described in EU countries, but the impact of different PEDV strains is difficult to compare between one country and another, as impact is dependent not only on pathogenicity but also on factors such as biosecurity, herd size, farm management, sanitary status or herd immune status.

• Increase awareness of clinical signs; high morbidity diarrhoea in pigs of any age is a good indicator of the need for PEDV testing.

• Biosecurity, in particular for vehicles, is important to prevent introduction of the virus on farm.

• Maintain vigilance as the EU pig herd is likely to be susceptible to novel PEDV variants.

• Link PEDV genome sequence data to epidemiological and outbreak investigation information in order to predict the contribution of insertion/deletion events to PEDV virulence and impact.

• Transparency and open communication concerning the occurrence of outbreaks of PED between farms and between countries is key for prevention of the spread of the virus and for effective response.

• Impact analysis requires a comparison with baseline values, where no disease is present and therefore more specialised studies might need to be conducted in order to clarify the disease impact in a farm. Appendix A -Tables summarising information from the extensive literature review Hokkaido (1) Tohoku (2) Kanto (3) Chubu (4) Kinki (5) Chugoku/Shikoku (6) Kyusyu/Okinawa (7) ]. Based on the analysis of this region, the nine strains were distributed between two major distinct clusters:

(1) a major clustersimilar to the original highly virulent US PEDV strains; and (2) similar to S INDEL US PEDV strain, OH851.

Cluster (1) 

263

Thailand (Eastern region) Two PEDV variants, designated CBR1 and EAS1, were isolated from3-day-old pigs Full-length genome analysis -CBR1 shares a high similarity ( Updated epidemiological data on PED www.efsa.europa.eu/efsajournal • Question 3: What is the impact on pig production of PED reported worldwide for pigs with laboratory confirmation of PEDmortality, morbidity and duration of disease in the four age classes (suckling, weaned, fatteners, adults).

In order to address these questions an extensive literature search will be performed in bibliographic databases, sources of grey literature and the web.

The literature search will be conducted using a range of relevant information sources, including searches in several relevant data bases.

Network members will also be asked to provide relevant country reports, grey literature, research studies or any other relevant information, provided that the information is allowed to be published in the EFSA Scientific Report.

Specific search strings, encompassing terminology related to Porcine Epidemic Diarrhoea (PED), will be used to search published literature (see Tables 1-5 ). The following bibliographic databases will be searched:

Web of Science, encompassing the following databases:

• 

PubMed Information will be also searched in ProMED-mail, Google Scholar and the OIE website, including the World Animal Health Information Database Interface (WAHID).

The final search string that was used for the database searches was: ((((((porcine epidemic OR PEDV OR PED))) OR diarrh*)) AND (swine OR pig OR porcine)). Concerning the Google Advanced Search the search string was porcine epidemic (diarrhoea OR diarrhea) filetype:pdf. The time period of publication was specified at the most detailed level available in each database. Therefore, when this was allowed, a detailed time period was specified as 1 October 2014 to 31 October 2015, while sometimes the period could only be specified as 2014-2015 and the actual publication date of the retrieved publications was accessed later during the process in order to assess their eligibility for inclusion in the review.

The output from the searched databases, including all indexed fields per hit (e.g. title, authors, abstract), will be exported into separate Endnote 7 files, allowing a count of the individual hits per database. Files will then be combined and duplicate records will be removed.

The files obtained will be transferred into DistillerSR Web-Based Systematic Review Software (Evidence Partners, Ottawa, Canada). Using the Distiller duplicate detection function, duplicates will be identified and removed (quarantined).

www.efsa.europa.eu/efsajournal TITLE: ((((((porcine epidemic OR PEDV OR PED))) OR diarrh*)) AND (swine OR pig OR porcine)) OR TOPIC: ((((((porcine epidemic OR PEDV OR PED))) OR diarrh*)) AND (swine OR pig OR porcine)) Timespan Eligible data include both individual and aggregated data that address the objectives of the review questions illustrated above. Information sources are: peer-reviewed journals; country reports that can become publicly available; dedicated web pages; publicly available laboratory reports, conference abstracts and presentations (that include enough information to allow the appraisal of the data reported), grey literature, unpublished research studies for which the researchers agree to make their data available to be included in the report and any other source of relevant information that can be identified.

Studies need to have been published or the information to have become available between October 2014 and October 2015, however, the information in the studies may concern previous time periods as well.

The studies may describe PED occurrence in any type of pig farming setting and for any pig ages. Selection criteria 1) Report relates to farmed domestic pigs (Yes = Include) and/or 2) Report indicates that laboratory testing was used to confirm PED infection (to include also serological information as optional) (Yes = Include) and/or 

Impact of PEDV outbreaks on pork production

Impact of porcine epidemic diarrhea on performance of growing pigs

Porcine epidemic diarrhea virus and discovery of a recombinant swine enteric coronaviruses

Investigating the introduction of porcine epidemic diarrhea virus into an Ohio swine operation

Study of Porcine Epidemic Diarrhoea Virus in UK Pigs at Slaughter in 2013. Animal Health and Veterinary Laboratories Agency

Complete genome sequences of two genetically distinct variants of porcine epidemic diarrhea virus in the Eastern Region of Thailand

Phylogenetic analysis of the spike (s) gene of the new variants of porcine epidemic diarrhoea virus in Taiwan

2015/10/31

AND

Acknowledgements: EFSA wishes to thank the members of the EFSA Network on PED: Adolf

3) Report is primary research or data collection (Yes = Include) and/or 4) Report describes (Yes = include) (a) prevalence, incidence, occurrence of the virus (b) full length or partial sequencing results (c) impact of the virus (morbidity, mortality, production losses, duration of disease) (d) vaccination and intervention strategies (e.g. biosecurity and bio-containment) (e) risk assessments for transmission between farms or between countries (e.g. feed and spray-dried porcine plasma (SDPP) used as a feed supplement)Reports including information on points d. and e. above will not be included in the current literature review but will be identified and kept in the database for future reference.All studies will be initially screened for their relevance to the review questions and allocated for inclusion in the literature review. Initial screening will be based on titles and abstracts and subsequent screening (for studies passing the first screening) will be based on the full article. Studies will be grouped by review questions.