Spatial analysis for bovine viral diarrhea virus and bovine herpesvirus type 1 infections in the state of Paraíba, northeastern Brazil

Abstract

Background: Bovine Viral Diarrhea Virus (BVDV) and Bovine Herpesvirus type 1 (BoHV-1) cause reproductive problems in cattle and restrictions on international trade in animals worldwide. Both infections were detected in cattle herds in the Paraíba state, Northeastern Brazil, however, the spatial distribution and geographic identification of positive herds for these viruses has never been examined. Therefore, the aim of this study was to describe the spatial pattern of apparent prevalence estimate and to identify spatial clustering of positive herds of BVDV and BoHV-1 infections in cattle herds from the state of Paraíba, Northeastern Brazil.

Results: The herd-level prevalence for BVDV and BoHV-1 infections in Paraíba were, respectively, 65.5% (95% CI: 61.1-69.7) and 87.8% (95% CI: 84.5-90.5). The average apparent within-herd prevalence of BVDV was 31.8% and of BoHV-1 was 62.4%. The predicted prevalence was highest (0.42-0.75) for BVDV in the west, north and eastern part of Sertão and in the central and eastern part of Agreste/Zona da Mata. For BoHV-1, the highest predicted prevalence (0.74-0.97) was in some local areas across Sertão and throughout the eastern part of Agreste/Zona da Mata. Six significant clusters were detected for BVDV, a primary cluster covering the eastern Sertão region, with 11 herds, radius of 24.10 km and risk relative (RR) of 2.21 (P < 0.001) and five smaller significant clusters, involving one or two herds in Agreste/Zona da Mata region with a high RR. A significant clustering of BoHV-1 positive herds (P < 0.001) was detected in Agreste/Zona da Mata region with a radius of 77.17 km and a RR of 1.27, with 103 cases. Consistency was found between kriging and SatScan results for identification of risk areas for BVDV and BoHV-1 infections.

Conclusions: The clusters detected contemplated different areas of the state, with BVDV cluster located in the Sertão and BoHV-1 in Agreste/Zona da Mata stratum. Through the risk mapping, it was possible to identify the areas in which the risk is significantly elevated, coincided with areas where there are borders with other states and in which there is a high movement of animals.

Keywords: BVDV; BoHV-1; Cattle; Cluster analysis; Epidemiology; Kriging estimate.

Fig. 1

Spatial localization of cattle herds in the State of Paraíba, Northeastern Brazil, according to sampling stratum. Detail shows Paraíba state within Brazil

Fig. 2

Spatial localization of positive and negative herds in the State of Paraíba, Northeastern Brazil: a Herd status of BVDV; b Herd status of BoHV-1

Fig. 3

Apparent within-herd prevalences for BVDV in the State of Paraíba, Northeastern Brazil: a) Kriging surface of predicted apparent prevalence and (b) the variance of kriging estimates of predicted apparent prevalence

Fig. 4

Apparent within-herd prevalences for BoHV-1 in the State of Paraíba, Northeastern Brazil: a Kriging surface of predicted apparent prevalence and (b) the variance of kriging estimates of predicted apparent prevalence

Fig. 5

Sample semivariograms with fitted model (solid line) used to generate kriging surface of predicted within-herd prevalence of BVDV (a) and BoHV-1 (b)

Fig. 6

Significant clusters of herds with a high apparent within-herd prevalence of BVDV (primary cluster: red, secondary clusters: yellow) and BoHV-1 (green)