Seroprevalence of hepatitis E virus infection among institutionalized psychiatric patients in Taiwan

1. Introduction 

Hepatitis E virus (HEV) is a positive, single-stranded RNA virus (Krawezynski, 1993) that is transmitted through fecal-oral route or consumption of contaminated water or food (Krawezynski, 1993, Velazquez et al., 1990, Huang et al., 1995, Wong et al., 1980). It can cause acute hepatitis and lead to high mortality in pregnant woman (Khuroo et al., 1981). The prevalence of hepatitis E infection was higher in developing and underdeveloped countries than in developed countries (Krawezynski, 1993, Velazquez et al., 1990, Huang et al., 1995, Wong et al., 1980, Lok et al., 1992, Hyams et al., 1992).

Taiwan is an endemic area of hepatitis B virus (HBV) infection, but has never had any outbreak of HEV infection. It has been reported that the prevalence of HEV infection in healthy adults was 10.7% in Taiwan (Lee et al., 1994). HEV also accounts for 1–2% of sporadic acute viral hepatitis and 4% of sporadic acute non-A, non-B hepatitis (Chu et al., 1999). Previous reports showed that acute sporadic hepatitis could occur in patients without any history of travel to areas where hepatitis E was endemic (Hsieh et al., 1998, Mansuy et al., 2004). Transmitted routes in these patients were obscure. Previous reports showed that swine handlers had high seroprevalence of HEV infection (Hsieh et al., 1999). Molecular analysis also revealed that isolates from swine and humans in Taiwan shared as high as a 97.3% identity and formed a distinct branch divergent from other strains of HEV worldwide (Wu et al., 2000). These evidences supported the swine might probably be a reservoir of HEV in Taiwan. In addition, inapparent routes may exist and cause subclinical infection.

Institutionalized patients are at high risk of suffering from infectious diseases, such as hepatitis A (Szmuness et al., 1977) and B (Koplan et al., 1978, Krugman et al., 1979). Among psychiatric patients, mongoloid patients often become hepatitis B surface antigen (HBsAg) carriers (Sutnick et al., 1972). Our previous study also showed the prevalence of anti-HCV was higher in the institutionalized psychiatric patients than in general population, although these patients contracted hepatitis B as the general population in Taiwan (Chang et al., 1993). However, the seroprevalence of anti-HEV in institutionalized psychiatric patients has not been reported before. In this study, 754 psychiatric patients were investigated for prevalence of HEV infection. The results were also correlated to clinical characteristics of these patients.

2. Methods 

3. Results 

A total of 754 institutionalized psychiatric patients (510 males, 244 females) were enrolled in the study. Their ages ranged from 19 to 76 years (median 42.0 years) and had been admitted to this institution for 0.1–31.8 years (median 8.3 years). The main psychiatric diagnoses were schizophrenia (96.4%), affective disorders (2.1%), mental retardation (0.9%), alcoholism (0.3%), and others (0.3%). These patients have been diagnosed with these disorders between 2 and 46 years (median 19.0 years).

Anti-HEV, anti-HCV, and HBsAg were positive in 109 patients (14.5%), 54 patients (7.2%), and 136 patients (18.0%), respectively. Anti-HEV positive patients were older than those anti-HEV negative patients (45.0±9.4 years versus 42.8±10.2 years, P=0.042). The males had higher prevalence than females (16.5% versus 10.2%; OR, 1.73; 95% CI, 1.07–2.78; P=0.023) (Table 1). As stratified by the patient's age, prevalence of HEV infection increased significantly by age group, with ORs of 4.89 (95% CI, 1.15–20.82; P=0.018) for 31–40 years old, 6.30 (95% CI, 1.48–26.83; P=0.005) for 41–50 years old, and 6.20 (95% CI, 1.44–26.74; P=0.006) for 50 years or older. Anti-HEV positive patients as compared to anti-HEV negative patients had similar duration of psychiatric disorders (20.7±9.3 years versus 19.6±8.6 years, P=0.233) and duration of hospitalization (8.9±6.0 years versus 8.6±5.4 years, P=0.598). Among these institutionalized psychiatric patients, HEV infection did not correlate with HBV or HCV infection, previous blood transfusions, and previous operations. In multiple logistic regression analysis, age and male gender were still the independent factors associated with HEV infection, with ORs of 1.021 (95% CI, 1.001–1.042; P=0.044) and 1.714 (95% CI, 1.058–2.777; P=0.029), respectively (Table 2).

Table 1. Characteristics of subjects and their seroprevalence of hepatitis E virus infection

	Variables	Number of subjects	Seropositive, n (%)	Odds ratio	95% confidence interval	P-value
	Sex
	Female	244	25 (10.2)
	Male	510	84 (16.5)	1.73	1.07–2.78	0.023
	Age (years)
	19–30	64	2 (3.1)	1
	31–40	279	38 (13.6)	4.89	1.15–20.82	0.018a
	41–50	231	39 (16.9)	6.30	1.48–26.83	0.005a
	≥51	180	30 (16.7)	6.20	1.44–26.74	0.006a
	HBsAg
	Negative	606	88 (14.5)
	Positive	136	18 (13.2)	0.90	0.52–1.55	0.698
	HCV Ab
	Negative	700	97 (13.9)
	Positive	54	12 (22.2)	1.78	0.90–3.49	0.092
	History of blood transfusion
	No	735	108 (14.7)
	Yes	18	1 (5.6)	0.34	0.05–2.60	0.276
	History of operation
	No	650	93 (14.3)
	Yes	103	16 (15.5)	1.10	0.62–1.96	0.742

a Data were compared with 19–30 year olds.

Table 2. Multivariate logistic regression analysis for factors associated with HEV infection

	Variables	Odds ratio	95% confidence interval	P-value
	Age	1.021	1.001–1.042	0.044
	Gender (M/F)	1.714	1.058–2.777	0.029
	HBsAg positive	0.931	0.536–1.616	0.800
	Anti-HCV positive	1.723	0.847–3.507	0.133
	History of blood transfusion	0.323	0.040–2.586	0.287
	History of operation	1.075	0.582–1.984	0.818

4. Discussion 

The study investigated HEV infection of institutionalized psychiatric patients in Taiwan. A significant percentage (14.5%) of those patients had HEV infection. The prevalence seems to be higher than those reported from other studies of various populations in Taiwan, including 10.7% of healthy individuals (Lee et al., 1994), 8.8% of volunteer blood donor and college students (Peng et al., 1995), 3.4% of preschool children, and 12.5% of kindergarten teachers (Lin et al., 2004). As comparing with the report from Peng et al., their studied population is similar to ours by the geographic location (southern Taiwan) and age distribution. Significantly higher prevalence of anti-HEV was present in institutionalized psychiatric patients when compared with healthy individuals aged more than 25 year-old (109/742 versus 44/485, P=0.0035), volunteer blood donors (109/754 versus 19/281, P<0.001), and vegetable farmers (109/754 versus 34/410, P<0.001) (Peng et al., 1995). HEV infection is usually transmitted by fecal contaminated drinking water or food. Possible factors which attributed to higher prevalence of HEV infection were found in underdeveloped and developing countries. The factors were a poor water supply system, ineffective sewage disposal, and easy contact with wild animals or their feces. In addition to documented transmitted routes, zoonosis is suggested as a possible contributing factor by several studies (Wu et al., 2000, Tei et al., 2003, Tei et al., 2004). According to previous reports, the Taiwan HEV strain from acute sporadic hepatitis is distinct from strains isolated from other parts of world (Hsieh et al., 1998). Moreover, the novel swine HEV strain in Taiwan shared high homology to the human strain (Hsieh et al., 1999). These results suggested that transmission of swine to human did occur. In Taiwan, pork is the major source of animal protein and, therefore, HEV transmission through swine to human in these institutionalized patients might occur. Swine handlers and dealers had higher HEV prevalence (Hsieh et al., 1999). Physical contact with infected subjects might have resulted in HEV transmission through inapparent routes. Institutionalized patients living in a crowded environment with poor sanitation and hygiene conditions probably increased the opportunity to spread the exposure of HEV. All of these factors possibly contributed to the high infection rate of hepatitis E in institutionalized psychiatric patients.

Our study showed that the prevalence of anti-HEV significantly increased with age. Only 3.1% of patients 30 years or less had previous HEV infection, but increased up to 13.6% for 31–40 year olds and reached its peak with those older than 41 years. The higher prevalence of hepatitis E infection among older populations was also observed previously in Taiwan (Lin et al., 2004, Lin et al., 2000) and Hong Kong (Wong et al., 2004). Higher HEV prevalence with increasing age in psychiatric patients reflects cumulative exposure to the virus during their lifetime and additional transmission routes related to the patients’ inappropriate behaviors or the environment of institution that could enhance the risk of infection. In addition, the health and hygiene environment in Taiwan improved with economic progress during the lifetime of patients under the age of 30 years. It could possibly be another reason to have a low prevalence rate of HEV infection in this younger population. However, it still needs further investigation and longitudinal follow-up study.

Interestingly, the study showed males have a higher prevalence of HEV infection than females. Although it is hard to have a precise reason to illustrate this finding, females do have superiority over males in some viral infections. Females have a higher remission rate and less probability of evolving severe chronic liver diseases after acquiring hepatitis B (Chu et al., 1993). In addition, our previous study showed that GB virus-C/hepatitis G virus infection more easily recovered in females than males (Cheng et al., 2000). Anti-HEV could persist 3–8 years after infection (Lee et al., 1994) and some decay later on. We, therefore, suspect the ability of females to recover faster than males from viral infections infers possible early decay of anti-HEV after infection. Furthermore, active physical activity, careless personal hygiene, and different life styles possibly contribute to a higher prevalence of HEV in males. Further study needs to illustrate males as one risk factor of getting HEV infection.

Although studies from Japan showed that HEV infection could be transmitted by parenteral routes and thereby induce HEV antibody positivity (Mitsui et al., 2004, Khuroo et al., 2004). Our study did not find HEV infection was associated with HBV and HCV infection, or any risk behavior related to parenteral transmission routes such as operations and blood transfusions. Definitely, HEV infection is transmitted mainly through fecal-oral transmission route.

In conclusion, institutionalized psychiatric patients have a higher prevalence of HEV infection than healthy subjects in Taiwan. The infection increased with age and was acquired more frequently in males than females. Subclinical infection might exist in these patients. For public health promotion, institutionalized psychiatric patients should be regarded as a specific group of HEV infection.