Epidemiology of vesicular stomatitis viruses in the southwestern United States

Abstract

The general objectives of these studies were to investigate the persistence of vesicular stomatitis viruses (VSV) in the southwestern United States, to identify management or environmental factors associated with exposure of horses to VSV in the southwest, and to investigate the applicability of sentinel herds for epidemiologic studies and for integration into animal disease surveillance systems. A longitudinal study of sentinel equine operations (SEO) was conducted in Colorado and New Mexico from May 1998 to October 2001. Visits were made approximately every 100 days to each SEO in which between 2 and 20 horses were given examinations of the mouth, nasal cavity, coronary bands and external genitalia. In addition, blood samples were drawn by jugular venipuncture and swabs collected from the oral cavity. Information about operation management practices, animal movement history and other potential risk factors was collected by standardized questionnaire at each visit. All blood samples were tested by a competitive enzyme-linked immunosorbent assay (cELISA) for antibodies to each serotype of VSV. Samples positive by either cELISA were further tested by a serum neutralization test and IgM capture ELISA. There were 537 (27.1%) and 889 (44.8%) positive cELISA tests out of 1,984 total tests for the New Jersey (VSV-NJ) and Indiana serotypes (VSV-IN) of VSV respectively. A total of 48 and 89 seroconversions occurred to VSV-NJ and VSV-IN respectively. A mixed model was developed to examine the association of operation-level and animal-level factors and the change in serum neutralizing antibody titers between consecutive SEO visits. Questionnaire data and data collected through the use of geographic information systems were used in the model. The ecological section and plant hardiness zone were ecological variables that were associated with changes in antibody titer to VSV-IN. There were no factors associated with changes in antibody titer to VSV-NJ. Climate data including daily maximum temperature, daily minimum temperature, daily mean temperature, daily mean relative humidity and daily total precipitation were collected at the Sevilleta Long Term Ecological Research site in central New Mexico. Discriminant analysis was used to identify the climatic variables best able to classify the months VS would occur between 1989 and 1999. The study found the amount of precipitation occurring 2, 10, 11, and 12 months previous to the month in which the cases were diagnosed were the climatic variables that best described the occurrence of VS cases. As an extension of this study, remote sensing data (normalized difference vegetation indices [NDVI]) along with climatic data were collected in relation to SEO in Colorado. The associations of the NDVI and climatic variables with the occurrence of cases of VS or seroconversions to either VSV-NJ or VSV-IN were investigated through discriminant analysis and logistic regression. A 10-period lag in total precipitation and 11-period lag in NDVI were variables that best predicted the occurrence of VS on Colorado sentinel equine operations. The occurrence of YS was investigated in El Salvador through monthly visits to 12 sentinel cattle operations located in four different departments. Management, environmental and spatial data were collected. Heifers were enrolled on the operations and were examined and bled monthly for 3 years. Two cELISAs were used to detect antibodies on each sample for each serotype of VSV. Small terrestrial rodents were trapped on 8 of 12 operations. The seroprevalence of VSV-NJ was found to be higher than the seroprevalence for VSV-IN. It was confirmed that VSV is endemic in the four departments investigated in El Salvador.