Department of Veterinary Population Medicine, University of Minnesota, 1365 Gortner Avenue, St. Paul, MN, USA.
Prev Vet Med. 2020 May;178:104977. doi: 10.1016/j.prevetmed.2020.104977. Epub 2020 Mar 25.
Analyses of livestock movement networks has become key to understanding an industry's vulnerability to infectious disease spread and for identifying farms that play disproportionate roles in pathogen dissemination. In addition to animal movements, many pathogens can spread between farms via mechanisms mediated by spatial proximity. Heterogeneities in contact patterns based on spatial proximity are less commonly considered in network studies, and studies that jointly consider spatial connectivity and animal movement are rare. The objective of this study was to determine the extent to which movement versus spatial proximity networks determine the distribution of an economically important endemic virus, porcine reproductive and respiratory syndrome virus (PRRSV), within a swine-dense region of the U.S. PRRSV can be classified into numerous phylogenetic lineages. Such data can be used to better resolve between-farm infection chains and elucidate types of contact most associated with transmission. Here, we construct movement and spatial proximity networks; farms within the networks were classified as cases if a given PRRSV lineage had been recovered at least once in a year for each of three years analyzed. We evaluated six lineages and sub-lineages across three years, and evaluated the epidemiological relevance of each network by applying network k-tests to statistically evaluate whether the pattern of case occurrence within the network was consistent with transmission via network linkages. Our results indicated that animal movements, not local area spread, play a dominant role in shaping transmission pathways, though there were differences amongst lineages. The median number of case farms inter-linked via animal movements was approximately 4.1x higher than random expectations (range: 1.7-13.7; p < 0.05, network k-test), whereas this measure was only 2.7x higher than random expectations for farms linked via spatial proximity (range: 1.3-5.4; p < 0.05, network k-test). For spatial proximity networks, contact based on proximities of <5 km appeared to have greater epidemiological relevance than longer distances, likely related to diminishing probabilities of local area spread at greater distances. However, the greater overall levels of connectivity of the spatial network compared to the movement network highlights the vulnerability of pig populations to widespread transmission via this route. By combining genetic data with network analysis, this research advances our understanding of dynamics of between-farm spread of PRRSV, helps establish the relative importance of transmission via animal movements versus local area spread, and highlights the potential for targeted control strategies based upon heterogeneities in network connectivity.
对牲畜流动网络的分析已成为了解行业对传染病传播的易感性以及确定在病原体传播中起不成比例作用的农场的关键。除了动物的移动，许多病原体还可以通过空间接近度介导的机制在农场之间传播。在网络研究中，基于空间接近度的接触模式的异质性不太常见，并且很少有同时考虑空间连通性和动物运动的研究。本研究的目的是确定动物运动与空间接近网络在确定美国猪密集地区一种具有经济重要性的地方性病毒——猪繁殖与呼吸综合征病毒（PRRSV）分布方面的程度。PRRSV 可以分为许多系统发育谱系。此类数据可用于更好地解析农场之间的感染链，并阐明与传播最相关的接触类型。在这里，我们构建了运动和空间接近网络；如果给定的 PRRSV 谱系在三年内每年至少有一次在网络中的一个农场中被回收，则网络中的农场将被归类为病例。我们评估了六个谱系和亚谱系在三年内的情况，并通过应用网络 k 检验来评估每个网络的流行病学相关性，以统计评估网络内病例发生的模式是否与通过网络连接的传播一致。我们的结果表明，尽管谱系之间存在差异，但动物运动而不是当地传播在塑造传播途径方面起着主导作用。通过动物运动相互关联的病例农场的中位数数量大约是随机预期的 4.1 倍（范围：1.7-13.7；p<0.05，网络 k 检验），而通过空间接近度相互关联的农场的这一衡量标准仅比随机预期高 2.7 倍（范围：1.3-5.4；p<0.05，网络 k 检验）。对于空间接近网络，<5km 的距离的接触似乎比更长的距离具有更大的流行病学相关性，这可能与距离较大时当地传播的可能性降低有关。然而，与运动网络相比，空间网络更高的整体连通水平突出了猪群通过该途径广泛传播的脆弱性。通过将遗传数据与网络分析相结合，这项研究提高了我们对 PRRSV 农场间传播动态的理解，有助于确定通过动物运动与本地传播进行传播的相对重要性，并突出了基于网络连通性异质性的有针对性的控制策略的潜力。
