Bluetongue Epidemiology in the European Union

Central and northern Europe are now at risk from bluetongue virus.

BT is a World Organization for Animal Health reportable disease and is of considerable socioeconomic concern and of major importance in the international trade of animals and animal products (4). Before 1998, BT was considered an exotic disease in Europe with just a few sporadic incursions (e.g., Spain and Portugal from 1956 through 1960) (6).
Our aim in this article is to provide a synthesis and some perspectives of BT epidemiology in the European Union (EU) since BTV's introduction in 1998. To this effect, we provide a short overview of the epidemiologic situation in Europe, followed by a brief description of the susceptible species, a discussion of the vectorial capacity and competence of the Culicoides spp. vectors, and an outline of the modes of introduction and mechanisms of amplifi cation.

Epidemiologic Situation in Europe BTV in EU, 1998-2005
During this 8-year period, at least 6 BTV strains belonging to 5 serotypes (BTV-1, BTV-2, BTV-4, BTV-9, and BTV-16) have been continuously present in parts of the Mediterranean Basin, including several member states of the EU (Table, Figure 1) (1,5,(7)(8)(9)(10)(11)(12). This emergence of BT into parts of Europe never before affected was attributed mainly to climate change and was linked to the northern expansion of the major Old World vector Culicoides imicola (Kieffer), which is an Afro-Asiatic species of biting midge (13). Additionally, novel indigenous European vector species of Culicoides within the Obsoletus and Pulicaris complexes were involved.
In the Mediterranean Basin 2 epidemiologic systems seem to predominate. The fi rst one is located in the eastern part of the basin, where serotypes 1, 4, 9, and 16 were identifi ed. In this system, the BTV strains originated in the Near, Middle, or Far East. The vectors included other species of Culicoides in addition to C. imicola. This fi nding was deduced from the fact that the disease penetrated into areas where C. imicola does not occur (the Balkans and beyond) (9). The involvement of novel vectors was subsequently confi rmed when the causative virus was isolated from mixed pools of 2 species, C. obsoletus (Meigen) and C. scoticus (Downes and Kettle), collected in central Italy (14) and from C. pulicaris (Linnaeus) in Sicily (15). The second epidemiologic system comprises the western part of the Mediterranean Basin, where serotypes BTV-1, BTV-2, BTV-4, and BTV-16 were identifi ed and the main vector is C. imicola. Although the appearance of BTV serotype 16 in this C. imicola system is the result of the westward spread of the virus across Europe (16), it is of particular interest because of strong indications that the fi eld virus may represent a reversion to virulence of the attenuated vaccine (e.g., in Corsica and in Sardinia in 2004, strains of BTV-16 isolated from the fi eld were identical to the live attenuated monovalent vaccine strain) (1,11) (Table, Figure 1).

BT in Central and Northern Europe, mid-August 2006 to late December 2007
BT was fi rst identifi ed in northern Europe in August 2006 and can be defi ned as an emergent disease in this zone (17). Between the date of the fi rst report (August 17,2006) and February 1, 2007 (18), 2,122 BT cases were entered into the European Commission's Animal Disease Notifi cation System (ADNS) (http://ec.europa.eu/food/animal/diseases/adns/index_en.htm) ( Figure 2) (19). In this region, in 2006, a pool of 50 nonengorged, parous C. dewulfi (Goetghebuer) in the Netherlands were positive by PCR for BTV (20), and several pools of C. obsoletus complex in Germany (i.e., not identifi ed down to species) were also PCR positive for BTV (21) (Figure 3). Although isolation of live BTV was not attempted in either instance, this research, conducted in an area where C. imicola does not occur, confi rms the earlier fi ndings of Mellor and Pitzolis, who isolated infectious BTV from nonengorged parous C. obsoletus in Cyprus, and shows that indigenous European Culicoides species can support a BT epizootic (22). Because C. obso-letus complex midges and C. dewulfi occur widely across central and northern Europe, this entire area must now be considered to be at risk for BTV (23,24).
Moreover, in relation to the demonstrated overwintering ability of the virus in northern Europe, small numbers of adult Culicoides spp. were captured in animal housing during the winter period (November 25, 2006, to March 9, 2007) (i.e., females of C. obsoletus complex, males of C. obsoletus, C. scoticus, and C. dewulfi ) (25). Whether the occurrence of these midges and the possibility of their activity extending over the winter in such climatically protected locations can explain the persistence of virus from 1 vector transmission season to the next (13) or whether they represent newly emerged midges from nearby breeding sites is not known (25). Several hypotheses have been formulated to explain the overwintering ability of BTV: by persistence 540 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 14, No. 4, April 2008 ‡This is an insufficiently attenuated vaccine strain (11). §This strain is indistinguishable from Onderstepoort BTV-2 live attenuated vaccine strain (7). within surviving adult vectors themselves, transovarial transmission through the vector, or prolonged/persistent infection in viremic or aviremic vertebrate hosts (13,25,26). The focus of interest now is to see if BTV is able to survive regularly between vector seasons and become endemic to northern Europe. The recrudescence of BTV-8 in northern France, the Netherlands, Belgium, Luxembourg, and Germany in 2007, and also the emergence of BTV-8 in the United Kingdom, Denmark, Switzerland, and the Czech Republic, suggests that this may well be the case (27,28). Unlike farther south, where populations of the traditional vector, C. imicola, peak in the late summer and autumn, when most BT cases occur, populations of the indigenous European vectors peak earlier in the year; whether this will be refl ected in a change in the temporal occurrence of BT cases remains to be seen. In the period from January 1, 2006, through December 28, 2007, 12 EU member states and Switzerland reported BT outbreaks on their territories, comprising all of the serotypes reported in Europe since 1998 ( Figure 4) (29,30).

Susceptible Species
BTV is transmitted between its ruminant hosts almost exclusively through the bites of the females of vector species of the Culicoides biting midge (31). The global distribution of BTV, therefore, is restricted to those regions where these vector species of Culicoides occur, and its transmission period is limited to the times when adult vectors are active. Depending on the species, adult vector activity generally starts some time in spring. Activity is positively correlated with temperature and reaches a maximum between 28°C and 30°C; activity decreases when the temperature drops and, for the traditional Afro-Asiatic vector C. imicola, is probably nonexistent at temperatures <10°C (13,31).
BTV can infect a broad spectrum of domestic and wild ruminants. However, serious clinical signs have been observed only in certain breeds of sheep (improved breeds) and a few deer species (32,33). Cattle and goats usually exhibit subclinical infections and therefore may serve as important and covert viral reservoirs for sheep (32). However, some serotypes such as serotype 8, which recently caused infection in northern Europe, exhibit a more important virulence in cattle (34,35) with serious socioeconomic consequences (5).

Vector Capacity and Competence
Risk for BTV infection is linked closely to the presence of adult vector Culicoides spp. (31). Until recently, C. imicola was believed to be the only important vector of BTV in southern Europe, but it is now known that several, newly recognized vector species are also involved. Others may be identifi ed in the future.
Vector competence of an insect species and vector capacity of an insect population are important parameters in this respect (36). Vector competence is the (innate) ability of a vector to acquire a pathogen, maintain it, and successfully transmit it to a susceptible host (13). Vector competence may be determined in the laboratory by providing groups of insects of a particular species with blood meals of appropriate concentrations of virus and assessing infection and transmission rates. Vector competence is defi ned as the proportion of feeding insects that support virus repli-

cation and transmit virus after a suitable incubation period.
In situations where transmission is diffi cult to demonstrate because of the technical problems in refeeding "diffi cult" insects such as Culicoides spp., it has become established practice to assume transmission if virus can be recovered from the salivary glands. Vector capacity refers to the potential for virus transmission of an insect population and takes into account a range of insect, host, and environmental variables, including vector abundance, vector survival, biting and transmission rates, host preferences, and host abundances, under a range of external (e.g., bioclimatic) conditions. Vector capacity can be defi ned as the number of infective bites that an infected vector causes during its lifetime (usually 2-4 weeks in the case of vector species Culicoides) (36,37).
Determining the 2 parameters explained above is essential to accurately estimate vector transmission rates and predict whether BTV will become established in an area. Such detailed studies inevitably demand substantial fi nancial and scientifi c resources and require a multidisciplinary approach.

Modes of Introduction and Mechanisms of Amplifi cation
Introduction of BTV from 1 area into another can occur in 4 ways: through animal movement (domestic and wild ruminants) or animal product transport (semen, embryos); by infected vector Culicoides spp. carried by various living (plants, animals) or inanimate (airplanes, ships) means; through the active fl ight of infected vector Culicoides spp. (local propagation); and through passive fl ight of infected vector Culicoides spp. on the wind (responsible for longdistance dissemination).
Whether the virus becomes established in a new area depends upon the number and distribution of susceptible hosts, the duration and titer of the BTV viremia in the hosts, the vector capacity of the local vector population, and the ambient temperature. In essence, establishment depends upon a suffi cient number of vector Culicoides spp. becoming infected by feeding upon local viremic hosts, surviving long enough to ensure completion of the intrinsic incubation period (4-20 days, depending on ambient temperature), and transmitting the virus by bite to new hosts (13). The extrinsic incubation period is the interval between when a vector is infected and when it fi rst becomes capable of transmitting the BTV to a new host (38). These requirements for BTV establishment have clearly been fulfi lled in much of southern Europe, as BTV has survived there in many locations since the late 1990s.

Conclusions
The widespread recrudescence of BTV-8 infections in northern France, Belgium, the Netherlands, Luxembourg, and Germany in 2007 and the emergence of BTV-8 in the United Kingdom, Denmark, Switzerland, and the Czech Republic in the same year suggest that the requirements for BTV establishment may now also be fulfi lled in many more 542 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 14, No. 4, April 2008  northerly and central parts of Europe (in the absence of C. imicola). In addition, the radial extension of BTV-8 across Europe (including the jump across the English Channel) ( Figure 5) (39) increases the risk for an encounter between this serotype and others, particularly those that occur in the Mediterranean Basin (second epidemiologic system). BTV serotypes 1, 2, 4, and 16 have been identifi ed in this area, and the addition of a further serotype will considerably increase the potential for reassortment between these viruses ( Figure  6) (27,40). Indeed, the number of possible reassortments in the case of BTV, which has 10 segments, increases with the number of cocirculating serotypes (e.g., 1,024 for 2 serotypes [2 10 ] and 59,049 for 3 serotypes [3 10 ]) (4). Moreover, the phenomenon of reassortment has already been demonstrated during the 1998-2005 BTV outbreaks in Europe (5). Furthermore, in the southern epidemiologic system, C. imicola, the Afro-Asiatic vector of BTV, occurs in addition to the C. obsoletus complex. As the population abundance of C. imicola peaks later in the year than the Obsoletus complex, this means that virus may be transmitted for a much greater portion of the year.
With regard to prophylaxis, possibly the best strategic option for control of clinical BT outbreaks in the European endemic areas is vaccination of susceptible animals with inactivated vaccines to protect against disease and to exclude the possibility of reversion to virulence of the vaccine viruses and reassortment between vaccine and fi eld strains of the virus (4,5). Veterinary authorities and legislators throughout northern Europe would do well take note of these recent and considerable changes in the epidemiology of BT. Figure 6. Bluetongue virus (BTV) restriction zones in Europe, by serotype. The radial extension of BTV-8 across Europe increases the risk for an encounter between this serotype and other serotypes that occur in the Mediterranean Basin (second epidemiologic system, where serotypes BTV-1, BTV-2, BTV-4, and BTV-16 were identifi ed and the main vector is Culicoides imicola). This situation increases the risk for reassortment of individual BTV gene segments, and, in the more southerly areas, the period of vector activity is also likely to extend, leading to a longer BTV-8 season. In addition, BTV-1, which was fi rst identifi ed in sheep with clinical signs of BT in the south of the Iberian Peninsula in July 2007, has extended its range into northern Spain and southwestern France (Pyrénées-Atlantiques), since November 2007; this ongoing expansion is matter of major concern.