The production losses in 13 Danish herds infected with the new PRRSV-1 variant have been documented in the present report. Only data from 13 of the more than 33 infected herds were obtained. Although the reason for participating or refusing to participate is not known for all the contacted herds, there is no indication that the participating herds were not representative of all the infected herds; however, this cannot be documented.
The herds included in the study had received semen from the infected boar station in July 2019 and experienced outbreaks in the following weeks. Before the outbreak, the monitoring of PRRSV at the boar stations was based on antibodies and performed every second week. After the outbreak, the surveillance was changed and now the boar stations are monitored every week based on detection of PRRSV by PCR and antibodies every second week.
The PRRSV isolated from the boar station had never been detected in Denmark or elsewhere prior to July 2019 and was therefore considered to be a new PRRS1-variant. The new PRRS1-variant is a recombination between the vaccine strains included in the Unistrain PRRS vaccine (Hipra, Spain) and the Suvaxzyn PRRSV (Zoetis Animal Health, USA) in that the ORF 1, ORF 2 and part of ORF 3 are more than 99% identical to the Suvaxyn strain, and the rest of the genome is more than 99% identical to the Unistrain vaccine strain. The breakpoint is located in ORF 3 encoding the glycoprotein 3 (GP3) after nucleotide position 201, corresponding to amino acid 67 . It is not the first time that a recombinant of two modified live PRRSV vaccines has been documented (9). Only herds in which the virus shared >99% homology with the Zoetis vaccine strain in ORF 2 and >99% identity with the Unistrain strain in ORF 5 were regarded as being infected through semen with the new variant strain and were included in the study.
For ten of the herds, the health status reported by SPF-SuS  was negative for PRRS prior to infection. Due to the low number of herds included in the study, the production losses cannot be compared for herds that were positive or negative to PRRS before infection with the new variant. Based on the visual inspection of the data, there do not appear to be any differences in the severity between herds that were positive or negative prior to the outbreaks.
Records from VETSTAT  show that, prior to PRRS infection, only one of the PRRS-positive herds (K) used mass vaccination against PRRSV prior to the outbreak. Despite this vaccination of all sows, the herd recorded production loses after infection with the recombinant PRRSV. After the outbreaks, most of the herd owners chose to vaccinate all sows, some several times, with a modified live PRRS vaccine. Nine of the 13 herds used "Unistrain PRRS", since the new PRRSV-1 variant is almost identical to the Unistrain vaccine strain in ORF 5, which is considered to be the major target for neutralising antibodies . Again, due to the low number of herds included in the study, the production losses cannot be compared with regard to the selected PRRS vaccine, although, based on the visual inspection of the data, there do not appear to be any differences. Since all herds decided to use mass vaccination against PRRSV, it cannot be stated whether this use of vaccines decreased or increased the production loses.
The number of farrowings per week was affected by infection with the new PRRSV-1 variant, with 12 herds experiencing a reduction, which is expected after infection with PRRSV. One herd had a small increase in farrowings per week, which was probably due to the fact that relatively more sows were inseminated after infection in order to compensate for a potential decrease in farrowings as a result of PRRSV.
Since PRRSV affects both the survival of the piglets and the reproduction of the sow, it is still too early to analyse the overall effects of PRRSV-1 infection on the reproductive parameters (e.g. wastage days and farrowing percentage), and therefore only effects including farrowings/week, liveborns/week, stillborns/week, pre-weaning mortality and weaned pigs/litter were included in the analysis.
The closest study to the one reported here seems to be the Danish study from 2013, which included eight herds. Seven herds were previously PRRS-free, and, of these, three were infected with PRRSV-1 and four with PRRSV-2. One herd had previously been infected with PRRSV-1 and had experienced an acute outbreak after introduction of PRRSV-2. The period with reduced production was estimated to be between ten and 90 weeks .
All 13 herds experienced fewer liveborn pigs at farrowing, which was also expected after an outbreak of PRRS. The average decrease in liveborn pigs per litter was 2.8 pigs for every litter during the five-month period after infection compared to the preceding seven months. When comparing this with the previous Danish estimate from 2013 , where the decrease was 0.7 liveborns/litter based on a seven-month period after infection with PRRSV, it is seen that this new PRRSV-1 variant results in higher production losses than older variants of PRRSV seen in Denmark. Even if it was assumed that the 13 herds went back to normal productivity for the last two months, so that the period after infection was seven months (as in the study from 2013), this new PRRS1-variant would still result in a decrease in liveborns of more than two liveborn pigs per litter.
In addition to there being fewer liveborn pigs, there was also an average increase in stillborn pigs per litter of 1.4 for the five months after infection compared to the preceding seven months. In the 2013 study, there was an average increase in stillborns per litter of just 0.3 during the seven-month period after infection with PRRSV. Thus, infection with the new PRRS1-variant appears to result in one more stillborn pig per litter compared to the previous outbreaks with other variants.
The pre-weaning mortality increased to an average of 40 percent after infection with the new PRRSV-1 variant. The increased pre-weaning mortality was also reflected in a lower number of weaned pigs from each litter in all 13 herds. On average, the 13 herds were missing 4.1 pigs from each litter at weaning when the five-month period after infection was compared to the preceding seven months. In the 2013 study, the eight herds were missing an average of only 1.0 weaned pig from each litter. Again, this new variant seems to have increased the negative impact on production parameters compared to other PRRSV variants previously seen in Denmark.
A reduction in liveborns per litter of between one and two pigs has been reported in several studies after infection with PRRSV [12, 13, 14, 15, 16, 17]. Similarly, an increased pre-weaning mortality of 4 to 17% [12, 13, 14, 16, 17, 18] and a decreased farrowing rate [19, 20, 21] have been reported from outbreaks in other countries. As a result of this decline in farrowings and increased piglet mortality, the number of weaned pigs per litter decreased by two to three pigs [12, 15, 20]. Differences in the virulence of circulating viruses, the presence of both PRRSV-1 and PRRSV-2 and differences in production systems and general health status make it difficult to compare studies of production losses resulting from the introduction of PRRSV in different regions/countries. Comparisons are even more difficult to make, since different countries have different ways of calculating and reporting data. Despite these limitations, it is clear that the losses seen in connection with this new recombinant PRRS1-variant exceeded not only the losses normally seen in Danish PRRSV-infected herds but also the losses seen in connection with PRRS outbreaks in other regions and countries.