After BTV-3 emerged in the Netherlands, and before any clinically suspicion cases had been announced in ruminants in Germany, we installed biting midge traps in animal stables in western Germany to collect putative BTV-3 vectors and test them for virus infection. Traps were equipped with a UV light but no CO₂ source (Biogents, https://eu.biogents.com). During September 24–26, we set 1 trap each on 18 cattle, sheep, and goat farms in North Rhine-Westphalia and Lower Saxony, along the border with the Netherlands. We placed the traps close to the animals at sites protected from wind and rain. Traps operated continuously, and we recovered collected insects every day until November 9 or 11, depending on the location. After those dates, the traps were only activated for 24 hours per week and samples were collected that day.

We collected biting midges, placed them in 80% ethanol and stored them at room temperature in the dark until processing. A few days, but not >4 weeks, after collection, we morphologically identified midges as C. obsoletus group, C. pulicaris complex, and other Culicoides. C. obsoletus group and C. pulicaris complex midges are considered the main BTV and SBV vectors in Europe (9). We used a multiplex quantitative reverse transcription PCR (qRT-PCR) to screen pools of <50 C. obsoletus group and C. pulicaris complexmidges for BTV and EHDV RNA (10). EHDV only recently emerged in southern and western Europe (7) and might be the next biting midge–borne virus to spread to central Europe. We also tested midge pools for SBV RNA because that virus is enzootic in the ruminant population in the region and expected to be detectable in insect vectors (11). We subsequently analyzed pools that tested BTV-positive by using a BTV-3–specific qRT-PCR (12). We retrospectively examined BTV-3–positive pools to determine the specific biting midge species (13,14).

During September 26–November 9, we collected 1,603 biting midge pools at 9 sites. The number of pools per site ranged from 27–466, depending on the number of midges collected. We tested those pools for viral RNA; 1 pool of C. obsoletus group midges collected in Kleve (Figure, panel C) on October 12 tested positive for BTV RNA (quantification cycle [Cq] value 35.6). We subsequently confirmed that pool as BTV-3–positive (Cq 37.5). The pool consisted of a mixture of C. obsoletus clade O1 (or C. montanus, which cannot be reliably differentiated from C. obsolestus clade O1 with the test system used but is not supposed to occur in Central Europe), C. scoticus, and C. chiopterus. Another pool of C. obsoletus group midges captured on the same day and at the same site tested SBV­ RNA–positive. In addition, we detected SBV in 534 midge pools collected during the 6.5-week period from all 9 locations: 1 site in Lower Saxony in the Grafschaft-Bentheim district, and 8 sites in North Rhine-Westphalia (3 in Kleve district, 2 in Wesel district, 2 in Borken district, and 1 in Heinsberg district) (Figure; Appendix). Except for 2 C. pulicaris complex pools, all SBV RNA–positive pools belonged to the C. obsoletus group (Appendix). No pools were positive for both BTV and SBV RNA, and all tested pools were EHDV-negative.

We calculated the minimum infection rates (MIR; i.e., number of positive pools divided by number of tested pools, multiplied by 1,000) for each virus (15). We found an MIR of 333.13 for SBV, indicating high circulation, and an MIR of 0.62 for BTV-3, indicating low circulation. However, MIR can be affected by pool size; the more specimens in a pool, the higher the possibility that >1 positive biting midge would be included, but that effect would not become evident when pools are examined. Conversely, the sensitivity of detection decreases with increasing pool size if only 1 positive biting midge was in the pool.

Using an isolate obtained from a BTV-3–positive sheep blood sample and further characterized on both Culicoides cells and baby hamster kidney cells, we produced a nearly complete genome sequence (International Nucleotide Sequence Database Collaboration, http://www.insdc.org; project no. PRJEB72862). The obtained genome was 99.94% identical to the sequence of a recent BTV-3 isolate from the Netherlands (GenBank accession nos. OR603992–4001) at the nucleotide level and 99.95% at the amino acid level. The genome segments of the strain from Germany were 83.24%–97.67% identical to the BTV-3 SAR2018 strain isolated from a sheep in Italy in 2018 (GenBank accession nos. MK348537–46) and 81.26%–97.89% identical to the TUN2016 strain isolated from a sheep in Tunisia in 2016 (GenBank accession nos. KY432369–78).

BTV-3 was confirmed in an infected sheep in Germany on October 12, 2023, and viral spread was detected in 2 federal states by winter 2023–2024. The isolated BTV-3 is nearly identical to virus strains from outbreaks in the Netherlands. A pool of C. obsoletus group biting midges collected on a cattle farm in the same district on the same day BTV-3 was confirmed in Germany tested positive for BTV-3 RNA. Detecting BTV-3 in its putative vectors confirms an ongoing transmission cycle, albeit circulating at a low level; only 1 insect pool tested positive, and only a few animals were BTV-3–positive on affected farms. In contrast, we found SBV RNA in numerous Culicoides pools, reflecting its intense circulation in ruminant populations; SBV-infected cattle, sheep, and goats have been reported in Germany since 2011, although prevalence between years varies (6). The Cq values of the SBV qRT-PCR in some of the investigated Culicoides pools indicate substantial virus loads, reflecting extensive regional SBV circulation in autumn 2023 (Appendix).

In conclusion, circulation of BTV-3 in Germany is likely to continue, intensify, and spread with the onset of seasonal biting midge activity in spring 2024. Large-scale biting midge monitoring combined with rapid analysis for viruses could contribute to an early warning system for emerging biting midge­–borne diseases.