Epidemiology and spatio‐temporal analysis of West Nile virus in horses in Spain between 2010 and 2016. Issue 2 (16th October 2017)

• Record Type: Journal Article
• Authors: García‐Bocanegra, I.; Belkhiria, J.; Napp, S.; Cano‐Terriza, D.; Jiménez‐Ruiz, S.; Martínez‐López, B.
• Is Part Of: Transboundary and emerging diseases. Volume 65:Issue 2(2018)
• Page Start: 567
• Page End: 577
• Publication Date: 2017-10-16
• Subjects: emerging disease -- MaxEnt -- maximum entropy -- risk‐based surveillance -- SaTScan -- West Nile disease; Veterinary medicine -- Periodicals; 636.089
• DOI: 10.1111/tbed.12742
• Languages: English
• ISSNs: 1865-1674

Abstract

Summary: During the last decade, West Nile virus (WNV) outbreaks have increased sharply in both horses and human in Europe. The aims of this study were to evaluate characteristics and spatio‐temporal distribution of WNV outbreaks in horses in Spain between 2010 and 2016 in order to identify the environmental variables most associated with WNV occurrence and to generate high‐resolution WNV suitability maps to inform risk‐based surveillance strategies in this country. Between August 2010 and November 2016, a total of 403 WNV suspected cases were investigated, of which, 177 (43.9%) were laboratory confirmed. Mean values of morbidity, mortality and case fatality rates were 7.5%, 1.6% and 21.2%, respectively. The most common clinical symptoms were as follows: tiredness/apathy, recumbency, muscular tremor, ataxia, incoordination and hyperaesthesia. The outbreaks confirmed during the last 7 years, with detection of WNV RNA lineage 1 in 2010, 2012, 2013, 2015 and 2016, suggest an endemic circulation of the virus in Spain. The spatio‐temporal distribution of WNV outbreaks in Spain was not homogeneous, as most of them (92.7%) were concentrated in western part of Andalusia (southern Spain) and significant clusters were detected in this region in two non‐consecutive years. These findings were supported by the results of the space–time scan statistics permutation model. A presence‐only MaxEnt ecological niche model was used to generate a suitability map for WNV occurrence in Andalusia. The most important predictors selected by the Ecological Niche Modeling were as follows: mean annual temperature (49.5% contribution), presence of Culex pipiens (19.5% contribution), mean annual precipitation (16.1% contribution) and distance to Ramsar wetlands (14.9% contribution). Our results constitute an important step for understanding WNV emergence and spread in Spain and will provide valuable information for the development of more cost‐effective surveillance and control programmes and improve the protection of horse and human populations in WNV‐endemic areas.