A wide range of arthropod-borne viruses threaten both human and animal health either through their presence in Europe or through risk of introduction. Prominent among these is West Nile virus (WNV), primarily an avian virus, which has caused multiple outbreaks associated with human and equine mortality. Endemic outbreaks of West Nile fever have been reported in Italy, Greece, France, Romania, Hungary, Russia and Spain, with further spread expected. Most outbreaks in Western Europe have been due to infection with WNV Lineage 1. In Eastern Europe WNV Lineage 2 has been responsible for human and bird mortality, particularly in Greece, which has experienced extensive outbreaks over three consecutive years. Italy has experienced co-circulation with both virus lineages. The ability to manage this threat in a cost-effective way is dependent on early detection. Targeted surveillance for pathogens within mosquito populations offers the ability to detect viruses prior to their emergence in livestock, equine species or human populations. In addition, it can establish a baseline of mosquito-borne virus activity and allow monitoring of change to this over time. Early detection offers the opportunity to raise disease awareness, initiate vector control and preventative vaccination, now available for horses, and encourage personal protection against mosquito bites. This would have major benefits through financial savings and reduction in equid morbidity/mortality. However, effective surveillance that predicts virus outbreaks is challenged by a range of factors including limited resources, variation in mosquito capture rates (too few or too many), difficulties in mosquito identification, often reliant on specialist entomologists, and the sensitive, rapid detection of viruses in mosquito pools. Surveillance for WNV and other arboviruses within mosquito populations varies between European countries in the extent and focus of the surveillance. This study reviews the current status of WNV in mosquito populations across Europe and how this is informing our understanding of virus epidemiology. Key findings such as detection of virus, presence of vector species and invasive mosquito species are summarized, and some of the difficulties encountered when applying a cost-effective surveillance programme are highlighted.

A human outbreak of West Nile virus (WNV) infections occurred in 2010 in central Macedonia, northern Greece. Most cases were observed close to four rivers forming a large Delta, a major Mediterranean wetland. WNV lineage 2 sequences were obtained from two pools of Culex pipiens mosquitoes trapped in sites where encephalitis cases occurred a few days before the trapping. The Greek strain showed the highest homology to Hungarian and South African strains, differing from the Russian WNV lineage 2 strain, which suggests that at least two lineage 2 strains have been introduced and established in Europe, causing severe disease to humans.

a b s t r a c tIn the summer of 2010 an epidemic of West Nile virus (WNV) occurred in Central Macedonia, Greece, with 197 human neuroinvasive disease (WNND) cases. In the following years the virus spread to new areas, with a total of 76 WNND cases in 2011, and 109 WNND cases in 2012 (14 and 12 WNND cases, respectively, in Central Macedonia). We established a surveillance system based on serological testing of domestic pigeons, using cELISA confirmed by serum neutralization test. In Central Macedonia, pigeon seroprevalence was 54% (95% CI: 49-59%) and 31% (95% CI: 24-37%) at the end of the 2010 and 2011 epidemic seasons, respectively. One serum was positive for neutralizing antibodies directed against Usutu virus. Pigeon WNV seroprevalence and incidence rates of human WNND after the 2010 epidemic were positively correlated ( = 0.94, at the regional unit level), while in 2011 the correlation ( = 0.56) was not statistically significant, possibly due to small number of human WNND cases recorded. To evaluate the efficacy of the system at alerting upon WNV enzootic circulation before the onset of human cases, we tested 270 pigeons in 2011 and 240 pigeons in 2012. In Central Macedonia, the first seroconversions in pigeons were recorded 44 and 47 days, respectively, before the first human WNND cases. Pigeon surveillance was used successfully for identification of areas with WNV enzootic transmission and for early warning. Timely diffusion of information to health authorities facilitated the implementation of preparedness plans to protect public health.

The individual filtering rates of the cladocerans in Creteil Lake were measured in the daytime with 14C-labeled ChZorella during a seasonal survey. This mesotrophic, shallow, polymictic lake is characterized by small algae (~25 wrn) and cladocerans (< 1.3 mm). Multiple regression models were established for each genus and for all the cladocerans. Body length alone explained from 44 to 57% of the total variance in the filtering rates of Daphnia spp., Ceriodaphnia spp., and Diaphanosoma brachyurum. An additional 23-34% of the variance was attributable to temperature. The inclusion of the Chl a concentration finally yielded r2 values ranging between 0.79 and 0.84. On the other hand, body length and temperature explained only 16% of the total variance in filtering rate of Bosmina longirostris. By taking into account the effect of factors other than length of the animal, the fit of the model established for all cladocerans improved considerably (from r2 = 0.47 to r2 = 0.83). Species-specific responses and thermal effects in the lake show the difficulty of applying models based solely on body length to obtain sufficiently accurate estimates of cladoceran filtering rates.