West Nile Virus Surveillance in 2013 via Mosquito Screening in Northern Italy and the Influence of Weather on Virus Circulation

Abstract

West Nile virus (WNV) is a recently re-emerged health problem in Europe. In Italy, an increasing number of outbreaks of West Nile disease, with occurrences of human cases, have been reported since 2008. This is particularly true in northern Italy, where entomological surveillance systems have been implemented at a regional level. The aim of this study was to use, for the first time, all the entomological data collected in the five regions undergoing surveillance for WNV in northern Italy to characterize the viral circulation (at a spatial and temporal scale), identify potential mosquito vectors, and specify relationships between virus circulation and meteorological conditions. In 2013, 286 sites covering the entire Pianura Padana area were monitored. A total of 757,461 mosquitoes were sampled. Of these, 562,079 were tested by real-time PCR in 9,268 pools, of which 180 (1.9%) were positive for WNV. The largest part of the detected WNV sequences belonged to lineage II, demonstrating that, unlike those in the past, the 2013 outbreak was mainly sustained by this WNV lineage. This surveillance also detected the Usutu virus, a WNV-related flavivirus, in 241 (2.6%) pools. The WNV surveillance systems precisely identified the area affected by the virus and detected the viral circulation approximately two weeks before the occurrence of onset of human cases. Ninety percent of the sampled mosquitoes were Culex pipiens, and 178/180 WNV-positive pools were composed of only this species, suggesting this mosquito is the main WNV vector in northern Italy. A significantly higher abundance of the vector was recorded in the WNV circulation area, which was characterized by warmer and less rainy conditions and greater evapotranspiration compared to the rest of the Pianura Padana, suggesting that areas exposed to these conditions are more suitable for WNV circulation. This observation highlights warmer and less rainy conditions as factors able to enhance WNV circulation and cause virus spillover outside the sylvatic cycle.

Fig 1. WNV and USUV detections.

Surveyed area with sampling stations and locations of virus detection (red, WNV-positive; azure, USUV-positive; black, WNV- and USUV-positive) and borders of the WNV (red line) and USUV (azure line) circulation areas. On the right: volume contour (VC) of the WNV-circulation area (red) and of the USUV-circulation area (blue) estimated by kernel density estimation (KDE). The plain area and plain limit are reported in green.

Fig 2. WNV- and USUV-positive polls and human cases of West Nile neuroinvasive disease (WNND).

Mosquito pools tested positive for WNV, USUV, and WNND cases, with reference to the sampling period and region (Emilia-Romagna, red; Lombardia, green; Piemonte, yellow; Veneto, blue; and Friuli Venezia-Giulia, azure; black line represent the mosquito minimum infection rate).

Fig 3. Nonlinear discriminant analysis (NLDA) and human cases of West Nile neuroinvasive disease (WNND).

Localization of WNND cases (black dots) and NLDA analysis showing the modeled distribution probability of WNV circulation (see legend for probability) in northern Italy according to the results from 78 surveyed stations. The red lines represent the 95% and 50% volume contour of the WNV-circulation area estimated by kernel density estimation (KDE).