Use of Hematological Parameters to Predict Disease Resilience of Pigs in Commercial Environments

• Author / Creator

  Lim, Jiehan

• High health status in nucleus and multiplication farms at the top of swine breeding pyramid is a barrier to genetic selection of disease resilient pigs. This resulted in a gap between the performance measured in the breeding company and the performance observed at the commercial level. At this time practical tools to select for resilient pigs do not exist. The overall objective of the thesis was to identify key differences in the complete blood count (CBC), namely blood erythrocyte, leukocyte and platelet concentrations between resilient and susceptible animals before and after natural challenge and to assess the potential of complete blood count (CBC) as a tool for breeders to select disease resilient animals in a high health environment.
  We studied 893 high health crossbred (Landrace x Yorkshire) barrows from multiplication farms. These barrows were introduced in batches and exposed naturally to multiple diseases simultaneously in a test station. Natural disease challenge was established using seeder pigs to simulate high disease pressure typical of a commercial situation. Performance traits (growth rate and treatment rate) were assessed and used to classify pigs into resilient and susceptible groups. Profiling of 29 haematological parameters was performed on whole blood samples collected from individual pigs before and after pathogen challenge in the natural challenge model.
  In chapter 3, in contrast to resilient pigs, susceptible pigs were found to have a low neutrophil concentration before challenge and a persistently higher concentration of neutrophils after disease challenge. This result was supported by the persistently higher expression of inflammatory cytokine genes such as TNF-α (P=0.08), IL-8 and IL-1B (P=0.09) in susceptible pigs after disease challenge. Although differences in red blood cells, platelets and their related traits before and after challenge did not provide any significant insights for the resilience trait, hemoglobin (FDR=0.05), MCV (FDR=0.02), together with WBC (lymphocytes (FDR=0.04) and monocytes (FDR=0.04)) were found to be significantly different between groups before challenge. These CBC traits were therefore potential predictors of resilience at the commercial level in high health farms. This result led to the study in chapter 4, where principal component analysis was used to reduce the dimensions of the CBC data before challenge. A linear prediction model was then trained using the generated principal components and stepwise feature selection. In this study, different resilient/susceptible classification methods were used and resulted in different proportions of animals in these two groups. Among these methods and their respective prediction models, method B yielded an average 55.0% prediction accuracy that was significantly higher (P<0.004) than prediction accuracy of random classifier, 50.0%. Although the accuracy is relatively poor the predicted resilient group in method B showed significantly lower treatment rate (P=0.05) and shorter days to market (P<0.05) that looked promising in terms of its practical use in the field.
  The findings from this thesis provide evidence that pre-challenge CBC could potentially be used by breeders in the classification of resilient and susceptible pigs in genetic nucleus and multiplication farms with high health status although further investigation is warranted to validate prediction results. Resilient and susceptible pigs showed different CBC and cytokine gene expression profiles before and after disease challenge. These differences may explain the disease/infection outcome of these two groups of pigs and could be developed as predictors of resilience.

• Subjects / Keywords

  • Disease Resilience
  • Complete Blood Count
  • Hematological
  • Prediction
  • Natural Disease Challenge

• Graduation date

  Fall 2018

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R34X54Z4F

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.