Department of Epidemiology, National Institute of Animal Health, 3-1-5, Kannnondai, Tsukuba, 305-0856, Japan

Department of Medical Biometry, University of Tübingen, Westbahnhofstr. 55-D, Tübingen, D-72070, Germany

Research Center for Tropical Infectious Diseases, Nagasaki University Institute of Tropical Medicine, Sakamoto 1-12-4, Nagasaki, 852-8523, Japan

Abstract

Background

Hepatitis E virus (HEV) infection is a zoonosis for which pigs play a role as a reservoir. In Japan, the infection has been enzootic in swine. Clarifying the detailed mechanisms of transmission within farms is required in order to facilitate an understanding of the age-specific patterns of infection, especially just prior to slaughter.

Results

Here we reanalyze a large-scale seroprevalence survey dataset from Japanese pig farms to estimate the force of infection. The forces of infection of swine HEV were estimated to be 3.45 (95% confidence interval: 3.17, 3.75), 2.68 (2.28, 3.14) and 3.11 (2.76, 3.50) [×10^{-2 }per day] in Hokkaido, Honshu and Kyushu, respectively. The estimates with our model assumptions indicated that the average ages at infection ranged from 59.0–67.3 days and that the basic reproduction number, _{0}, was in the order of 4.02–5.17. Sensitivity analyses of age-specific incidence at different forces of infection revealed that a decline in the force of infection would elevate the age at infection and could increase the number of virus-excreting pigs at the age of 180 days.

Conclusion

Although our estimates imply that more than 95% of pigs are infected before the age of 150 days, the model shows that a decline in the force of infection could increase the risk of pig-to-human transmission. If the force of infection started to decline, it might be necessary to implement radical countermeasures (e.g. separation of uninfected pigs from infected herds beginning from the end of the suckling stage) to minimize the number of virus-positive pigs at the finishing stage.

Background

Hepatitis E virus (HEV) is a positive-strand RNA virus without an envelope, which is classified as a member of the genus

HEV infection in humans is seen not only in developing countries but also in industrialized countries where sporadic cases of infection have been reported

With these points in mind, it is essential to clarify the detailed mechanisms of HEV transmission in swine. For example, it would be very useful to know the average age of individuals acquiring infection in enzootic areas and the age-specific incidence, especially just prior to slaughter. Moreover, to identify effective control measures on the farm (e.g., potential vaccination strategy _{0}, defined as the average number of secondary cases arising from a single primary case in a fully susceptible population. _{0 }gives an indication of the transmission potential, and thus, is one of the most important epidemiologic determinants _{c}, can be derived by using _{0}; _{c }> 1-1/_{0 }_{0 }can be approximately obtained by estimating the force of infection (i.e. the rate at which susceptible individuals become infected),

The aim of this paper was to assess the transmission potential of HEV infection in swine using seroprevalence survey data from Japan. To clarify the age-specific mechanisms of transmission in swine raised for human consumption, published data

Results

Time required for seroconversion

A simple logit model was applied to the cumulative distribution of the time required for seroconversion of anti-HEV antibody based on inoculation experiments using swine HEV ^{2 }goodness-of-fit test did not reveal significant deviations between observed and predicted frequencies (^{2}_{5 }= 2.31, p = 0.81).

Observed and predicted time required for seroconversion of anti-HEV antibody

**Observed and predicted time required for seroconversion of anti-HEV antibody**. Logit model (line) was applied to the observed cumulative distribution of the time required for seroconversion (dot), which revealed a sigmoid pattern. Data source: ref. [9].

Estimates of the force of infection

The total sample sizes of seroprevalence surveys were 1400, 400 and 700 for Hokkaido, Honshu and Kyushu, respectively ^{-2 }per day]. Expected values of ^{-2 }day^{-1 }indicate that the average time of infection ranged from 29.0–37.3 days after the age of 30 days when individuals were first exposed to the risk of infection (i.e. average age at infection was 59.0–67.3 days). Observed and predicted age-specific seroprevalence are compared in Figure ^{2}_{4 }= 16.71, p < 0.01), this was not the case for Honshu (^{2}_{4 }= 1.49, p = 0.89) or Kyushu (^{2}_{4 }= 6.34, p = 0.17).

Observed and predicted age-specific seroprevalence against swine hepatitis E virus in Japan

**Observed and predicted age-specific seroprevalence against swine hepatitis E virus in Japan**. Observed (gray bar) and predicted (black) seroprevalence are compared. Three discrete geographic areas, Hokkaido (**A**), Honshu (**B**) and Kyushu (**C**), are modeled separately. Data source: ref. [24].

The population structure on pig farms in Japan is specific in that individuals are slaughtered immediately after the age of 180 days (i.e., 150 days after pigs are first exposed to the risk of infection). This satisfied a reasonable approximation of a simple age-specific survivorship function, referred to as Type I survivorship

where _{0}, is approximated as the product of the force of infection,

_{0 }≈

Thus, _{0 }(and the 95% CI) was estimated to be 5.17 (4.76, 5.62), 4.02 (3.43, 4.71) and 4.66 (4.13, 5.25) for Hokkaido, Honshu and Kyushu, respectively.

Age-specific incidence at different forces of infection

Figure ^{-1}). Accordingly, the average ages at infection,

Cumulative frequency of infection and age-specific incidence at different forces of infection

**Cumulative frequency of infection and age-specific incidence at different forces of infection**. **A**. Cumulative frequencies of HEV infection and **B**. age-specific incidence elicited by different forces of infection are compared. Assumed values for the forces of infection were 0.01 (thick black), 0.03 (thin black) and 0.05 (thick gray) days^{-1}. See eqs. 6 and 7 for details of the model.

Discussion

This study estimated the force of infection of swine HEV for three geographic locations in Japan. For the estimation, we incorporated two realistic aspects of swine HEV transmission: (1) no exposure during the suckling stage and (2) time delay of seroconversion after exposure to the virus. As a result, the force of infection was estimated to be approximately 0.03 day^{-1 }implying that the average age at infection is 63 days after birth. According to the estimates, the basic reproduction number, _{0}, was in the order of 4–5, which is relatively high compared to other diseases

There are two practical implications from our exercise. First, estimation of the force of infection permitted clarification of the average age at infection (being 63 days). Although our model did not allow more detailed age- and time-specificity of the force of infection to be derived due to limited data

Second, the critical coverage of vaccination required for eradication, _{c}, is obtained from _{0}, using _{c }> 1-1/_{0 }_{0}, ranging from 4.02–5.17, suggests that the HEV transmission on the farm could be prevented if more than 75.1–80.7 % of the pigs were successfully immunized. However, since HEV infection in man is likely to result in asymptomatic or mild disease

Conclusion

The force of infection of swine HEV was estimated from three discrete geographic locations in Japan using age-specific seroprevalence data. The estimates ranged from 2.68–3.45 ×10^{-2 }(day^{-1}), indicating that _{0 }ranges from 4.02–5.17. The estimates permitted a reasonable prediction of the age-specific incidence including that at the finishing stage. Although our estimates of the force of infection imply that more than 95% are infected before the age of 150 days and the probability of virus-excretion is small at 180 days, the model suggests that a decline in the force of infection could elevate the average age at infection and increase the risk of pig-to-human transmission. If the force of infection started to decline, it might be necessary to implement radical countermeasures (e.g. separation of uninfected pigs from infected herds beginning from the end of the suckling stage) to minimize the number of virus-positive pigs at the finishing stage. As this study showed a reasonable estimation in Japan which is an enzootic area for swine HEV infection, similar seroprevalence survey would be extremely useful to decipher the mechanisms of transmission. Thus, seroepidemiologic studies of swine, human and other animals with time, space and age as well as among specific groups

Methods

Data

To estimate the force of infection, this study used two published datasets: (1) an experimental inoculation study of swine HEV

Estimation of the force of infection of hepatitis E virus in Japan

**Estimation of the force of infection of hepatitis E virus in Japan**. **A**. The map of three geographic locations in Japan (drawn by the authors). **B**. Compartment of the catalytic model. Susceptibles at age **C**. Schematic illustration of the time delay to seroconvert. If the time of seroconversion _{0 }and possible time of exposure _{k }are given, probability of exposure at time _{k }can be extracted by _{0}-_{k}), where g(

Estimation of the time required for seroconversion

Cumulative distribution of the time required for seroconversion,

where

_{m})

where _{m }is the median time required for seroconversion. In order to apply a logistic curve to the cumulative distribution, the model has to satisfy _{m }were obtained by minimizing the binomial deviance of the model from the observed data. The 95% CI were determined by using the profile likelihood.

Force of infection

For simplicity and due to limited data availability, we assumed that the force of infection,

for

for

In addition to the estimation of

Convolution equation and maximum likelihood estimation

Since all individuals are slaughtered immediately after the age of 180 days, the time delay from infection to seroconversion was thought to be non-negligible. That is, the age-specific seroprevalence data (at age _{i}, whose infection is reflected as seropositive at

Using the same mid-point of the time-interval, the probability density of time to seroconvert, _{j}, at

The density of those who newly showed seropositive results at the age of 60 days (i.e., at _{1}, is given by _{1 }= _{1}g_{1}. In the same way, the densities at the ages of 90 and 120 days, _{2 }and _{3}, are given by _{2 }= _{2}g_{1 }+ _{1}g_{2 }and _{3 }= _{3}g_{1 }+ _{2}g_{2 }+ _{1}g_{3}. This can be generalized by using the following convolution equation

Since the observed seroprevalence data shows the cumulative distribution of those seroconverted after infection until month _{i}, which is given by

where _{i }pigs who had been seropositive until month _{i }who had not, the likelihood function is

The maximum likelihood estimate of

Abbreviations

HEV – Hepatitis E virus

CI – Confidence interval

Competing interests

The author(s) declare that they have no competing interests.

Authors' contributions

KS and HN carried out paper reviews, proposed the study, performed statistical analyses and drafted the manuscript together. The authors have read and approved the final manuscript.

Acknowledgements

We thank Prof. Hiroaki Okamoto, Jichi Medical School, for permitting us to reanalyze the seroprevalence data with model. HN thanks to the Banyu Life Science Foundation International for supporting his research in Germany. This study was also supported by the Japanese Ministry of Education, Science, Sports and Culture in the form of a Grant-in-Aid for Young Scientists (#18810024, 2006).