A farm‐to‐consumption quantitative microbiological risk assessment for hepatitis E in pigs

Foodborne transmission appears to be a significant route for human hepatitis E virus (HEV) infection in Europe. We have developed a quantitative microbiological risk assessment (QMRA) for HEV infection due to consumption of three selected pork products (liver pâté, minced meat, and sliced liver), which models the steps from farm to human consumption in high detail, including within-farm transmission dynamics and microbiological processes such as cross contamination and thermal inactivation. Our model is unique in that it considers prevalence and viral load of two microbiological variables, HEV RNA and infectious HEV, expressing the latter in terms of the former through so-called "adjustment factors" where data are lacking. When the QMRA was parameterized for France and using infectious HEV, we found that sliced liver posed by far the highest risk of infection, with mean probability per portion <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mn>3.35</mml:mn> <mml:mo>×</mml:mo> <mml:msup><mml:mn>10</mml:mn> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>4</mml:mn></mml:mrow> </mml:msup> <mml:mspace></mml:mspace> <mml:mrow><mml:mo>[</mml:mo> <mml:mn>95</mml:mn> <mml:mo>%</mml:mo> <mml:mspace></mml:mspace> <mml:mtext>CI</mml:mtext> <mml:mspace></mml:mspace> <mml:mrow><mml:mo>(</mml:mo> <mml:mn>3.28</mml:mn> <mml:mo>-</mml:mo> <mml:mn>3.42</mml:mn> <mml:mo>)</mml:mo></mml:mrow> <mml:mo>×</mml:mo> <mml:msup><mml:mn>10</mml:mn> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>4</mml:mn></mml:mrow> </mml:msup> <mml:mo>]</mml:mo></mml:mrow> </mml:mrow> <mml:annotation>$3.35\times 10^{-4}\,[95\%\ \text{CI}\ (3.28-3.42)\times 10^{-4}]$</mml:annotation></mml:semantics> </mml:math> , corresponding to <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mn>3447</mml:mn> <mml:mspace></mml:mspace> <mml:mo>(</mml:mo> <mml:mn>95</mml:mn> <mml:mo>%</mml:mo> <mml:mspace></mml:mspace> <mml:mtext>CI</mml:mtext> <mml:mspace></mml:mspace> <mml:mn>3372</mml:mn> <mml:mo>-</mml:mo> <mml:mn>3522</mml:mn> <mml:mo>)</mml:mo></mml:mrow> <mml:annotation>$3447\,(95\%\ \text{CI}\ 3372-3522)$</mml:annotation></mml:semantics> </mml:math> human cases annually. For minced meat, the probability of infection was <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mn>3.68</mml:mn> <mml:mo>×</mml:mo> <mml:msup><mml:mn>10</mml:mn> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>8</mml:mn></mml:mrow> </mml:msup> <mml:mspace></mml:mspace> <mml:mrow><mml:mo>[</mml:mo> <mml:mn>95</mml:mn> <mml:mo>%</mml:mo> <mml:mspace></mml:mspace> <mml:mtext>CI</mml:mtext> <mml:mspace></mml:mspace> <mml:mrow><mml:mo>(</mml:mo> <mml:mn>3.56</mml:mn> <mml:mo>-</mml:mo> <mml:mn>3.80</mml:mn> <mml:mo>)</mml:mo></mml:mrow> <mml:mo>×</mml:mo> <mml:msup><mml:mn>10</mml:mn> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>8</mml:mn></mml:mrow> </mml:msup> <mml:mo>]</mml:mo></mml:mrow> </mml:mrow> <mml:annotation>$3.68\times 10^{-8}\,[95\%\ \text{CI}\ (3.56-3.80)\times 10^{-8}]$</mml:annotation></mml:semantics> </mml:math> , with only <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mn>21</mml:mn> <mml:mspace></mml:mspace> <mml:mo>(</mml:mo> <mml:mn>95</mml:mn> <mml:mo>%</mml:mo> <mml:mspace></mml:mspace> <mml:mtext>CI</mml:mtext> <mml:mspace></mml:mspace> <mml:mn>20</mml:mn> <mml:mo>-</mml:mo> <mml:mn>21</mml:mn> <mml:mo>)</mml:mo></mml:mrow> <mml:annotation>$21\,(95\%\ \text{CI}\ 20-21)$</mml:annotation></mml:semantics> </mml:math> human cases. While our model predicted appreciable levels of HEV RNA remaining in liver pâté at the point of consumption, the amount of infectious HEV and hence risk of infection was zero, emphasizing the importance of using the correct microbiological variable when assessing the risk to consumers. Owing to its highly mechanistic nature, our QMRA can be used in future work to assess the impact of control measures along the pork-supply chain at high resolution.