The course of African swine fever in Romanian backyard holdings – A case report

Abstract African swine fever (ASF) was diagnosed for the first time in Romania in a backyard holding in Satu‐Mare County in July 2017. Since then, more than 3800 outbreaks occurred in the entire country. Disease control strategies in the backyard sector rely almost exclusively on reactive measures implemented upon appearance of clinical signs and laboratory confirmation of ASF. In our descriptive study, infection course and outbreak investigation data of 56 affected backyard holdings in Satu‐Mare County has been investigated. Early disease detection based on clinical signs appeared to be efficient. In the majority of outbreaks, ASF was detected within the first 2 weeks after the estimated virus introduction. A clinical phase of 2–8 days was observed before pigs either succumbed to the disease or control measures were implemented on affected farms. A moderate on‐farm transmissibility of ASF virus between pigs was observed. Four clusters of outbreaks were identified indicating virus perpetuation and transmission from farm to farm. To suspend infection chains, rapid intervention by isolating affected farms combined with effective biosecurity measures is required. However, due to the backyard peculiarities, quick and effective implementation of control measures has shown to be rather difficult.

Despite the low number of animals per farm, and the low relevance on global trade volumes, backyard holdings can play an important role in the local dynamics of ASF epizootics  and therefore impact the disease status of a country. A general feature observed in backyard holdings is insufficient biosecurity to prevent introduction and spread of ASF. Pigs in backyard holdings are often fed with kitchen leftovers, cereals and fresh grass. This kind of feed is prone to be contaminated with ASFV and thereby representing a high risk for disease introduction (Bellini et al., 2016). Although swill feeding is legally banned, in practice it is difficult to control (Boklund et al., 2020).
In experimental studies with ASFV genotype II strains, the average time between infection and death has shown to be around 10 days, rather independent from the virus dose applied (Pietschmann et al., 2015). This period includes an incubation time of around 5 days and the following clinical phase when pigs show clinical signs and shed virus. Particularly in the early clinical phase, clinical signs are rather unspecific and often mistaken for other infectious diseases. For effective disease control, the time between occurrence of clinical signs and isolation of the affected holdings should be as short as possible.
The timely isolation of the affected holding and culling of animals is likely to be a crucial factor for reducing virus transmission and secondary infections. The earlier interventions are taken the lower is the probability of virus transmission to other farms.
So far, there are not many reports published, dealing with ASF in backyard holdings outside of Africa (Boklund et al., 2020;Zani et al., 2019). Our study aims to describe main findings (morbidity, mortality and laboratory data) of ASF outbreaks in backyard holdings in Satu-Mare County. The data contributes to a better understanding of ASF in backyard settings, particularly regarding the course of the disease and the transmission patterns within and between holdings. This knowledge could help to improve regional strategies addressing ASF control in the backyard sector.

Definitions
Clinical phase: Time from the first reported appearance of clinical signs in pigs on the farm until death or culling.

RESULTS
From the very first case in July 2017 until 2019, 56 ASF outbreaks were notified in backyard holdings in Satu-Mare County (Table 1) In large farms, the very first animals falling sick and dying from ASF might be overlooked if there is no enhanced passive surveillance system in place targeting dead and sick animals with a specific follow-up (Lamberga et al., 2018). In the majority (86%) of the farms of our study, the estimated clinical phase lasted 2-6 days before measures (isolation of the backyard, culling of animals) were implemented. Hence, it can be hypothesised that on these farms ASF was detected towards the end of the first infection period of around 10 days (Gabriel et al., 2011). This hypothesis is supported by the fact that in 40% of the backyard holdings the disease was discovered before pigs succumbed to the disease.
On two backyard holdings, seropositive animals were found, indicating an infection period of more than 10 days (Gabriel et al., 2011).
Generally, ASFV-specific antibodies have been rarely reported from European countries affected with ASFV Genotype 2 as pigs mostly succumb to the disease before seroconversion. In other backyard holdings, clinical signs were noticed in different animals over a period of up to 31 days but no suspicion was raised. The observed variations can be explained with differences between holdings, unspecific clinical signs as well as with the subjectivity in reporting clinical observations. Swill is considered to be generally the most likely source of virus introduction in the backyard sector (Costard et al., 2015;Heilmann et al., 2020). In experimental studies, it was proven that the oral infectious dose may have an impact on the infection rate. To efficiently infect healthy pigs by the oral route, more than 10,000 infectious units of ASFV are needed (McVicar, 1984). However, with much lower doses, weaker animals will pick up the infection and a scattered infection pattern is observed in the herd (Pietschmann et al., 2015). Even with high infectious doses, some animals may not get infected and contagiousness can be moderate (Gabriel et al., 2011;Zani et al., 2018).
Nevertheless, it could be proven that liquids (e.g. contaminated water) containing low doses of virus are more infectious than contaminated forages (Niederwerder et al., 2019).
In our study, we observed a delayed disease transmission if pigs were kept in separate pens and different stables. This is in line with experimental studies, where it could be shown that an infectious pig would infect on average 5.0 animals within one pen and 2.7 animals between pens. The within-pen transmission might be facilitated by blood contact (high virus dose). Contaminated materials, for example contaminated stable equipment or clothes of workers might contribute to the in-between pen transmission (Guinat et al., 2016). A similar observation was made in an outbreak in a backyard holding in Bulgaria .
We identified four clusters of outbreaks (A, B, C, D). Due to their proximity in time and space, we assumed that the outbreaks within a cluster were epidemiologically linked. However, clear proofs for possible links between the farms were not found. Timelines of clustered outbreaks suggest that in some cases simultaneous infections of farms occurred leading then to secondary outbreaks. It can be assumed that the virus leaped from one backyard holding to the next due to insufficient biosecurity measures, for example frequent movements of people and shared equipment. The observed mortality in clustered outbreaks was found to be lower than in non-clustered outbreaks. This can be explained by the fact that after disease confirmation neighbouring farms were checked timely leading to earlier disease detection.
To facilitate early detection and to avoid secondary spreading, it is crucial that after an outbreak, all neighbouring backyard holdings are inspected immediately for the presence of ASF. This measure should be repeated every week, at least for 1 month after the last outbreak.
In particular, the presence of clinical signs and mortality should be checked. If suspect or dead animals are found, samples should be taken and tested for ASFV. Additionally, incentives for backyard farmers that report timely sick and dead animals could facilitate early detection. The success of such a reactive strategy is mainly depending on the rapid intervention by isolating affected farms combined with effective biosecurity measures.

CONFLICT OF INTEREST
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

ACKNOWLEDGEMENT
The authors would like to thank Felix Närmann for preparing the map in the supporting information.
Open Access funding enabled and organized by Projekt DEAL.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article and supplementary material; further inquiries can be directed to the corresponding author.

PEER REVIEW HISTORY
The peer review history for this article is available at https://publons.