Cyprinid herpesvirus 3 (koi herpesvirus, KHV) caused a number of outbreaks in the Pardubice region (Czech Republic) in the 2018–2020 period. The aim of this work was to investigate the origin and spread of the virus using conventional and molecular epidemiology, including partial genomic analysis of individual viruses. Koi herpesvirus disease (KHVD) first appeared in the region in 2018 in two sites. Despite veterinary measures, the virus returned to one of the affected ponds in 2019 and was detected in five additional locations in the region. In 2020, additional outbreaks of KHVD were detected in the region, this time with no apparent link to previous outbreaks. The analysed viruses were collected in the years 2018–2020 (Appendix S1). Host species was predominantly common carp (Cyprinus carpio), with one sample representing koi carp (Cyprinus carpio). With one exception, all viruses originated from the Pardubice region. Samples were collected by the Veterinary Administration authorities, either after reported clinical signs of KHVD or during active surveillance. The initial laboratory examination was conducted at the State Veterinary Institute in Jihlava (Czech Republic) and subsequent confirmation and virus characterization at the National Reference Laboratory for Viral Diseases of Fish at the Veterinary Research Institute in Brno, Czech Republic. At each of the positive locations, information about stocking and transports of live animals, purchases of fish, winter storage ponds and potential contacts of fish from various locations was provided. Additionally, the possibility of virus transport via fishing equipment and shared personnel was documented. On site, the health status of the fish was monitored, along with equipment and feed storage conditions and disinfection procedures. Watercourses in the area were mapped, emphasizing potential virus sources and ponds with the increased risk of disease introduction. Gill, spleen, heart and cranial kidney tissues from up to five animals were pooled and examined for the presence of KHV using real-time PCR (Gilad et al., 2004). To distinguish between Asian and European KHV strains, duplex PCR according to Bigarré et al. (2009) was employed. Specific viruses were additionally examined via conventional PCR, with partial SphI-5 (610 bp), 9/5 (611 bp) (Kurita et al., 2009) and thymidine kinase genes (1117 bp) amplified (Padhi et al., 2019; Appendix S3). The PCR mix was based on the GoTaq G2 Flexi Polymerase (Appendix S4; Pokorova et al., 2010). The PCR products were purified and sequenced using Sanger direct sequencing (SEQme). The forward and reverse-sense sequences were assembled in MEGA 6 (Tamura et al., 2013) and uploaded to the GenBank database (Appendix S2). Three main factors contributing to virus dissemination were determined: import and transport of fish, shared watercourses and shared fishing equipment (Table 1). Potentially positive locations were identified via either increased mortality and pathoanatomical signs (gill necrosis, skin changes), or as inapparent infections during surveillance measures. After laboratory confirmation, outbreak sites, along with high-risk negative ponds, were eradicated using pond depopulation, draining and disinfection with quicklime (3 t CaO/ha), followed by a 6-week fallowing period. The affected locations are marked in Figure 1. The first confirmed outbreak site in the Pardubice region was in 2018 in the main pond 1/a (CZ2129). A second KHVD outbreak of 2018 was confirmed in fingerling pond 2 (CZ2130), with very poor water-quality conditions. After a period of disinfection and fallowing, ponds 1/a and 1/b were re-stocked with fish from four fingerling ponds in the spring of 2019. An apparent KHVD infection broke out in all of the aforementioned ponds in summer of the same year (CZ2218, CZ2225, CZ2229, CZ2242, respectively), along with the main pond 1/a (CZ2215). In 2020, the first outbreak was registered in the private garden pond 8 (CZ2296) with ornamental koi fish. The site was not connected to the 2019 cases. The second KHV-positive location of 2020—Ponds 9/a (CZ2300) and 9/b—was freshly stocked in spring 2020, again without apparent connection to the 2019 outbreaks. Duplex (Bigarré et al., 2009) and conventional (Shahin et al., 2020) PCR were used to determine the virus lineage. Twelve viruses were of CyHV3-U/ I lineage. Only a single virus (CZ2296) originating from a private koi carp pond belonged to CyHV3-J lineage (Figure 2). Gene regions 9/5 and Sph I-5 (Kurita et al., 2009) proved to be too conservative to distinguish individual viruses (Appendices S5 and S6), but the partial thymidine kinase gene (Padhi et al., 2019) showed variability (Appendix S7). While the 2018–2019 viruses from the Pardubice region were 100% identical, the CyHV3-J virus CZ2296 was significantly variable (9 bp difference) and the virus CZ2300 collected in 2020 showed 1 bp variability, along with virus CZ2241 originating from outside of the Pardubice region. The initial introduction of KHV into the Pardubice region could not be reliably traced, but the most likely source of infection is the importation of infected fish in the autumn of 2017. According to the owner, carp stock in this period was predominantly imported from Supplier 1, forming a common denominator for most KHVD outbreaks. The imported fish showed no apparent signs of infection but were not laboratory tested for the presence of KHV. This, along with the ability of koi herpesvirus to cause latent infections (Eide et al., 2011) created an opportunity for the introduction of the virus. Pond 2 (CZ2130) was proposed as a source of initial clinical infection, with poor water quality possibly causing the reactivation of latent disease (Lin et al., 2017) due to increased stress levels in the summer of 2018 (McColl, 2013). After the initial introduction of KHV to the region, multiple ponds were infected, predominantly via the routine transport of live fish for winter storage and restocking of the main ponds. Ponds with separate fish stock were additionally tested or suspected positive due to a shared watercourse (e.g. ponds 1/b, 4/c) (Perelberg et al., 2003) and the use of non-disinfected fishing equipment (e.g. ponds 4/d, 5/b, 7/b, 7/c) (Way, 2016). Conversely, the disinfection of equipment with 5% chloramine probably prevented the reinfection of fresh stock from Supplier 2 in pond 4/a, despite the use of shared equipment with infected ponds 4/b – 4/d (Smail et al., 2004). The measures preventing reinfection of each pond: Depopulation, disinfection and fallowing (Flamm et al., 2016; EU, 2015/1554) were proven effective in all locations except for the pond 1/a. More than virus persistence, the reintroduction of the virus with infected fish stock from fingerling ponds tested KHV-positive in 2019 is likely in this case. Sequencing analysis of three possible targets on the KHV genome showed the SphI-5 and 9/5 genes (Kurita et al., 2009) to be too conservative, but a partial TK gene analysis (Padhi et al., 2019) to be sufficient to detect variability of individual viruses. Sequences from 2018 to 2019 collected in the Pardubice region were 100% identical in the 1117 bp sequence (main cluster). This observation is supported by conventional epidemiology data rooting the virus distribution in one source disseminated by fish transport. The sequence CZ2241 from outside of the Pardubice region diverged from the sequence of the main TK cluster, as did the two viruses studied in 2020 from the Pardubice region. This again supports the conventional epidemiological data, which failed to link the 2020 outbreaks to the 2018–2019 cases. Duplex and conventional PCR (Bigarré et al., 2009; Shahin et al., 2020) confirmed the Asian origin of the 2020 sample CZ2296 isolated from koi carp, in contrast to all of the other studied isolates assigned to European lineage. The diverging TK sequences of viruses CZ2241 and CZ2300 showed a small but significant rate of mutations for a double-stranded DNA virus in the relatively short time span of the study (Sanjuán et al., 2010) suggesting a more widespread underlying presence of KHV in Czech aquaculture. As only a few outbreaks of KHVD are usually reported in the country per year (European Commission, 2022) the role of inapparent and latent infections cannot be ruled out (Eide et al., 2011) underlining the importance of active surveillance programmes (EU, 2015/1554). Routine carp relocation, shared watercourses and unsanitized fishing equipment have allowed the spread of koi herpes virus with 100% phylogenetic identity to 10 ponds in the Pardubice region in 2018–2019. The spread of the virus was halted in 2019 using strict eradication measures: active surveillance, depopulation, draining, disinfection and fallowing of positive ponds. Additional 2020 outbreaks from the same region showed no epidemiological connection to the previous years, as confirmed by molecular examination. This work was supported by the Ministry of Agriculture of the Czech Republic MZE-RO0523 and Project PROFISH CZ.02.1.01/0.0/0.0/16_019/0000869 financed by ERDF in the operational program VVV MŠMT. The authors declare no conflict of interest. The data that supports the findings of this study are available in the supplementary material of this article. Appendix S1–S4 Appendix S5 Appendix S6 Appendix S7 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.