Prevalence of hepatitis E virus antibodies in pigs: implications for human infections in village-based subsistence pig farming in the Lao PDR

Summary

We report a high seroprevalence of hepatitis E virus (HEV) in pigs in the Lao PDR. HEV seroprevalence was 51.2% (300/586) amongst abattoir pigs and 15.3% (46/301) amongst village pigs. The age distribution suggested previous in-village HEV pig infections. These findings suggest a zoonotic risk associated with village-based smallholder pig farming.

1. Introduction

Hepatitis E virus (HEV) is an enterically transmitted zoonotic viral agent that causes acute jaundice in humans with evidence that pigs and other mammals play a role in maintenance and transmission. In mainland Southeast Asia, human HEV infections have been described in Vietnam and Thailand (Corwin et al., 1996; Meng et al., 1999) and recently in Lao PDR (Laos), where of 400 patients admitted to hospital in Vientiane with hepatitis or jaundice, 15.4% had evidence of past infection with hepatitis E and 7 (1.7%) had serology and/or PCR evidence for acute hepatitis E infection (P.N. Newton, unpublished results). The people of Laos are largely rural and reliant on village-based agriculture. In many villages pigs are an important source of food and cash. In this study we examined the seroprevalence of HEV antibodies in pigs in villages and at abattoirs throughout Laos to better understand the potential role of pigs in the transmission of HEV to humans.

2. Materials and methods

No animal samples were obtained specifically for this study. Pig sera from two archival collections that were originally collected for classical swine fever virus sero-surveillance were tested: (a) sera were collected from pigs (n = 1912) being slaughtered at abattoirs in 16 provinces (89% of the total of 18 provinces) between January and August 2001, and a subset was selected using random number tables (n = 586; 30.6%) aiming for 40 sera per province when possible (Table 1); (b) randomly selected pigs were venesected in villages in March 1998 in all nine districts of Vientiane Municipality, resulting in the sampling of 301 pigs (total n = 1902; 15.9%) from 50 randomly selected villages (Table 1). Samples were stored at −20°C until analysis.

Table 1. HEV seropositivity for pig sera collected at slaughter and in villages.

Province	Districta	Sampleb	Collection period	n	HEV positive villagesc n (%)	HEV Ig positive n (%)
Attapeu	N/A	A	Jan–Aug 2001	39	NR	14 (36.0)
Bhorikamxai	N/A	A	Jan–July 2001	40	NR	20 (50.0)
Bokeo	N/A	A	Jan–July 2001	40	NR	14 (35.0)
Champassack	N/A	A	Jan–Aug 2001	40	NR	10 (25.0)
Huaphan	N/A	A	May 2001	7	NR	6 (85.7)
Khammouane	N/A	A	Jan–Aug 2001	40	NR	22 (55.0)
Luang Namtha	N/A	A	Jan–Aug 2001	40	NR	25 (62.5)
Luang Prabang	N/A	A	Feb–June 2001	17	NR	8 (47.1)
Oudomxai	N/A	A	Jan–Aug 2001	40	NR	24 (60.0)
Phongsaly	N/A	A	Jan–Aug 2001	40	NR	16 (40.0)
Salavane	N/A	A	Jan–July 2001	40	NR	14 (35.0)
Savannakhet	N/A	A	Jan–July 2001	40	NR	30 (75.0)
Sekong	N/A	A	Jan–July 2001	40	NR	29 (72.5)
Vientiane Municipality	N/A	A	Jan–Aug 2001	40	NR	19 (47.5)
Xayabuly	N/A	A	Jan–July 2001	40	NR	18 (45.0)
Xieng Kouang	N/A	A	Jan–Aug 2001	43	NR	31 (72.1)
Total		A		586		300 (51.2)
Vientiane Municipality	Hadxaifong	V	March 1998	31	5 (83.3)	10 (31.3)
Vientiane Municipality	Pak Ngeum	V	March 1998	68	3 (33.3)	6 (8.8)
Vientiane Municipality	Xaithany	V	March 1998	36	2 (33.3)	5 (13.9)
Vientiane Municipality	Sikottabong	V	March 1998	55	2 (20.0)	7 (12.7)
Vientiane Municipality	Sisattanak	V	March 1998	22	2 (66.6)	3 (13.6)
Vientiane Municipality	Naxaithong	V	March 1998	22	2 (50.0)	2 (9.1)
Vientiane Municipality	Chantabuly	V	March 1998	31	4 (57.1)	7 (22.6)
Vientiane Municipality	Sangthong	V	March 1998	23	1 (33.3)	3 (13.0)
Vientiane Municipality	Xaysettha	V	March 1998	13	1 (50.0)	3 (23.1)
Total		V		301	23 (46.0)	46 (15.3)

^aN/A: not available.

^bA: abattoir pigs; V: village pigs.

^cNR: not relevant.

Detection of HEV Ig was performed with an ELISA (Innis et al., 2002), using a recombinant capsid protein antigen of HEV (rHEV) Sargodha 1987 strain. The positivity cut-off (>40WR U/ml of total Ig) was determined by calculating the mean WR U/ml ± 3 SD of 30 Thai HEV-negative pigs. Samples that were ELISA positive were confirmed by Western blotting using SDS-PAGE 12% pre-cast gel (NuPAGE Novex, Invitrogen, Carlsbad, CA, USA) and nitrocellulose WB strips, and were considered positive for rHEV capsid protein antibody if positive and negative controls gave expected results and a band was clearly visible at 56 kDa MW.

Data analysis was performed using STATA version 8 (Stata Corp., College Station, TX, USA). Sample HEV IgG prevalence was calculated as the proportion of HEV serologically positive pigs in the sampled population (village, district, province or age group) and significant differences between sample groups (P ≤ 0.05) were determined using Student’s t-test for paired samples and ANOVA.

3. Results and discussion

The average HEV seroprevalence for abattoir pig samples was 51.2% (300/586), with provincial prevalence rates ranging from 25.0% (Champassak) to 85.7% (Huaphan) (Table 1). Among village pigs in Vientiane Municipality, the overall HEV seroprevalence was 15.3% (46/301); district seroprevalence rates ranged from 8.8% (Pak Ngeum) to 31.3% (Hadx-aifong) (Table 1). HEV seropositive pigs were recognized in 23 villages (46.0%) in Vientiane Municipality. There were no significant differences in the HEV seroprevalence between provinces and districts (P>0.05).

The age-stratified seroprevalence suggests previous HEV infections in village-raised pigs. While passively acquired maternal antibodies would be detected up to approximately 3 months of age, the results suggest new HEV infections beyond 3 months with seroprevalence peaking at 7–9 months (Figure 1). Similar patterns of HEV age-related seropositivity have been described in other studies with peaks between 4 and 6 months of age (Meng et al., 1999; Wibawa et al., 2004) with lower seroprevalence rates in adult pigs (Meng et al., 1999). Age dependent differences in HEV seroprevalence rates may explain the higher seroprevalence in abattoir samples as these pigs would generally be from the 6–9 month age range. However, differences between village and abattoir samples may also be caused by other factors such as geographical locality, and farming and husbandry practises.

Figure 1. Comparison of HEV seropositivity in pigs of different age ranges.

The high level of pig HEV seropositivity, the suggestion of acute HEV infections in pigs, frequent mixing of pigs and humans, and poor sanitation in Lao villages suggests the potential for pig to human transmission (Wibawa et al., 2004). The common practice in Lao villages of fertilizing vegetables with pig manure and raising pigs over fishponds may provide additional sources for human HEV infection. Furthermore, the consumption of uncooked pig meat, a common practice in Laos, is recognized as a potential route of human HEV infection (Wibawa et al., 2004) and a source for Trichinella spiralis infection (Sicard et al., 1976). Raising the awareness of the importance of the adequate cooking of meat, the boiling of drinking water and washing of vegetables are likely to be effective public health interventions to prevent HEV infections. However, to conclusively establish a link between pig and human HEV infections in Laos further genetic and epidemiological studies are required.