Analysis of seroprevalence and risk factors for hepatitis E virus (HEV) in donation candidates and blood donors in Northeast Brazil

Abstract

Oral transmission is the main route of hepatitis E virus (HEV) infection; however, genotypes 3 and 4 may also be transmitted by blood transfusion. Individuals who need blood products are often immunosuppressed, which increase the risk of severe disease and death by HEV. Despite this, blood banks in Brazil do not screen for HEV and epidemiological studies in this population are rare; this is an important issue as HEV-3 is frequently identified in the country. Herein, we analyzed the seroprevalence and risk factors for HEV seropositivity in donor candidates/blood donors from Northeast Brazil. Nine hundred and ninety-six donor candidates/blood donors from Foundation of Hematology and Hemotherapy of Pernambuco (HEMOPE) were interviewed regarding socioeconomic, sociodemographic, and behavioral data and analyzed for anti-HEV IgG. Anti-HEV IgG was detected using the HEV IgG (EUROIMMUN) kit. Associations between seropositivity and potential risk factors were analyzed by the χ² test and Fisher’s exact test. Seroprevalence was 0.9% (9/996), 77.77% (7/9) and 22.22% (2/9) in blood donors and donor candidates, respectively. HEV seropositivity was associated with male (OR: 11.65; CI: 0.6755–200.9; p = 0.0163), income higher than BRL 20,000/month (p = 0.0002), and lake bathing (OR: 4.553; CI: 1.391–15.25; p = 0.0258). Importantly, about 43% (3/7) of anti-HEV positive donors made their first donation more than 20 years ago, which must be taken as a warning sign, given the possibility that these individuals may have been infected after registration as donors. Finally, the report of HEV seropositivity, especially in regular blood donors, as well as the identification of potential risk factors, reinforces the need for viral screening in Brazilian blood banks.

Introduction

Hepatitis E virus (HEV) (currently Paslahepevirus balayani) belongs to the genus Paslahepevirus and family Hepeviridae [1]. To date, eight HEV genotypes have been described (1–8), with genotypes 1, 2, 3, 4, and 7 related to human infection [2, 3]. Genotypes 1 and 2 are transmitted by drinking contaminated water, causing major outbreaks worldwide [4, 5]. Conversely, genotypes 3 and 4 are transmitted mainly through the consumption of raw/undercooked meat from infected animals (e.g., pigs, wild boar and deer), although unconventional transmission routes, such as the ingestion of contaminated shellfish and parenteral route, have also been recognized [6, 7, 8].

It has been reported that HEV 3 may remain viable in blood for 4 weeks after collection [8], enough time to contaminate blood products that may be destined for patients with specific clinical conditions, such as hemodialysis and transplant recipients, who are often also immunosuppressed [8, 9]. In immunosuppressed patients, infection with HEV genotype 3 may progress to a chronic infection with risk of liver damage, fibrosis, and cirrhosis [10, 11].

Given the importance and possibility of HEV transmission in blood banks, several studies have reported the detection of HEV RNA and/or anti-HEV antibody in blood donors [8, 12, 13, 14, 15, 16, 17]. In line with this context, the European Association for the Study of the Liver (EASL) has recommended screening for HEV at least for blood products intended for immunosuppressed patients [18]. The UK, Ireland, France, Netherlands, Germany, Spain, Austria, and Luxembourg have implemented an HEV screening program for blood donors, while other European countries are looking into this possibility [19].

In Brazil, studies on HEV in blood donors are rare. Recently, Tengan et al. [20], in a systematic review based on seven studies, reported a 7% seroprevalence for HEV in blood donors in the country. In addition to the limited number of studies, seroprevalence values differ considerably depending on the Brazilian region analyzed [20]. In Northern and Southern regions, for example, seroprevalence for HEV was found to be 0.44% and 40.25%, respectively [21, 22]. Regarding the Northeast, the second most populous Brazilian region, only one study, published in 1997, analyzed the HEV seropositivity in blood donors, finding a 2% seroprevalence [23]. The little attention given to HEV in blood donors becomes even more worrying when one considers that the epidemiological profile in Brazil is similar to that observed in developed countries, where HEV genotype 3 predominates [24].

Considering the aforementioned, and recognizing the importance of HEV for public health, specifically for patients who undergo blood transfusions and/or who use blood products, we describe the HEV seroprevalence in donor candidates and blood donors from Pernambuco state, Northeast Brazil, and analyze the factors of risk related to HEV seropositivity.

Material and methods

Design and population/local of study

This is a cross-sectional analytical study carried out with donor candidates and blood donors from the Foundation of Hematology and Hemotherapy of Pernambuco (Fundação de Hematologia e Hemoterapia de Pernambuco, HEMOPE) between April and November 2021. HEMOPE is a Brazilian reference center in transfusion medicine, assistance for organ transplant services, laboratory diagnosis, and treatment of blood diseases, located in Pernambuco state, Northeast Brazil. The Pernambuco state occupies an area of approximately 98,312 km² and is divided into four intermediate geographic regions: Recife, Caruaru, Serra Talhada, and Petrolina (see Fig. 1).

Fig. 1.

Spatial distribution of individuals screened and positive for anti-HEV IgG from HEMOPE. The Pernambuco state (A), located in Northeast Brazil (B), is divided into four intermediate geographic regions: Recife, Caruaru, Serra Talhada, and Petrolina, which have 72, 63, 25, and 25 municipalities, respectively (here delimited by black lines). The Foundation of Hematology and Hemotherapy of Pernambuco (Fundação de Hematologia e Hemoterapia de Pernambuco, HEMOPE) is located in Recife intermediate geographic region, specifically in the municipality of same name (Recife)

The study included individuals over 18 years old, regardless of serologic screening for human immunodeficiency virus (HIV), human T-cell lymphotropic virus (HTLV), hepatitis B or C virus (HBV or HCV), Chagas disease, or syphilis. After signing the Free and Informed Consent Term, the individuals were interviewed for the collection of sociodemographic, socioeconomic, and behavioral data. Regarding individuals who were blood donors, the date of the first and last donation, as well as the number of donations made to date, were obtained from the donor register. The study was approved by the Ethics and Research Committee of HEMOPE, protocol 4.636.584, and Certificate of Presentation of Ethical Appreciation 40,987,620.0.3001.5195.

Sampling calculation

The minimum sample for the study was 270 individuals. The sample size was determined by the Epi info program (version 6.04), with an alpha error of 5% and a confidence interval (CI) of 95%, and was based on the 2% prevalence of anti-HEV IgG in blood donors from Bahia state, Brazil [23]; to date, the only study available in Northeast Brazil for sample size calculation.

Detection of anti-HEV IgG

From everyone submitted to the interview, 5 mL of peripheral blood were collected. Blood samples were processed and stored at − 80 °C until use. Detection of anti-HEV IgG was performed by ELISA, using the Hepatitis E (HEV) IgG (EUROIMMUN) kit, according to the manufacturer’s instructions.

Statistical analysis

The association between HEV seropositivity and potential risk factors was assessed by univariate analysis using the χ² test and Fisher’s exact test. The p-value < 0.05 and 95% CI were considered in all analyses. Statistical analyses were performed in the GraphPad Prism program (version 9).

Results

During the study, 996 donor candidates/blood donors were interviewed, of which 304 were blood donation candidates (first time) and 692 were blood donors. The mean age was 32.89 (± 11.11) years old, and 621 (62.35%) and 375 (37.65%) were men and women, respectively.

The prevalence of anti-HEV IgG was 0.9% (9/996), and all individuals with positive serology were male, inhabitants of the Recife intermediate geographic region (Fig. 1), with a mean age of 34.22 (± 14.33) years old. Among the nine positive individuals, two were donor candidates and seven were blood donors. Regarding the latter, the dates of the first and last donation, as well as the number of donations made, are shown in Table 1 and include individuals who made up to 82 donations in the last almost 24 years. Among sociodemographic, socioeconomic, and behavioral data, male sex (OR: 11.65; CI: 0.6755 to 200.9; p = 0.0163), income above BRL 20,000/month (p = 0.0002), and lake bathing (OR: 4.553; CI: 1.391–15.25; p = 0.0258) were associated with HEV seropositivity (Tables 2 and 3).

Table 1.

Data from anti-HEV IgG positive blood donors identified in the study

Identification code	Number of donationsa	First donation	Last donation
A	82	12/24/1997	10/09/2021
B	24	12/24/1997	06/10/2021
C	10	09/17/1998	05/18/2021
D	10	02/21/2014	06/02/2021
E	2	11/11/2020	08/24/2021
F	2	01/18/2021	05/05/2021
G	1	04/22/2021	04/22/2021

^aNumber of donations made since the donor’s registration at the Foundation of Hematology and Hemotherapy of Pernambuco (Fundação de Hematologia e Hemoterapia de Pernambuco, HEMOPE)

Table 2.

Univariate analysis of sociodemographic and socioeconomic data of donor candidates and blood donors screened in the study

Variables	Anti-HEV (IgG) positive (%)	Anti-HEV (IgG) negative (%)	ORa (95% CIb)	p-value
Sexc				0.0163
Male	9/9 (100)	612/987 (62.01)	11.65 (0.6755 to 200.9)	-
Female	0/9 (0)	375/987 (37.99)	-	-
Aged				0.651
< 30	4/9 (44.45)	440/987 (44.58)	-	−
30–40	2/9 (22.22)	290/987 (29.38)	-	-
41–50	1/9 (11.11)	159/987 (16.11)	-	-
> 50	2/9 (22.22)	98/987 (9.93)	-	-
Ethnicityd,e				0.251
Black	5/9 (55.56)	512/987 (51.87)	-	-
White	1/9 (11.11)	313/987 (31.71)	-	-
Brown	3/9 (33.33)	162/987 (16.41)	-	−
Marital statusc				1.000
Single	5/9 (55.56)	587/987 (59.47)	0.851 (0.2564 to 2.793)	−
Married	4/9 (44.44)	400/987 (40.53)	-	-
Educationd
Illiterate	0/9 (0)	6/987 (0.61)	-	0.2142
Elementary school	3/9 (33.33)	110/987 (11.14)	-	-
High school	4/9 (44.45)	630/987 (63.83)	-	−
Higher education	2/9 (22.22)	241/987 (24.42)	-	-
Income/month (BRL, R$)d
0–2200	7/9 (77.78)	678/987 (68.69)	-	-
2201–4400	1/9 (11.11)	190/987 (19.25)	-	-
4401–10,000	0/9 (0)	98/987 (9.93)	-	-
10,001–20,000	0/9 (0)	17/987 (1.72)	-	-
> 20,000	1/9 (11.11)	4/987 (0.41)	-	0.0002
Piped waterc				1.00
Yes	9/9 (100)	946/987 (95.84)	0.833 (0.04764 to 14.57)	−
No	0/9 (0)	41/987 (4.16)	-	−
Use of well waterc				1.00
Yes	6/9 (66.67)	642/987 (65.05)	1.075 (0.2535 to 3.944)	-
No	3/9 (33.33)	345/987 (34.95)	-	−

^aOR: odds ratio

^bCI: confidence interval

^cCalculated by the Fisher’s exact test

^dCalculated by the χ² test

^eEthnic classification was based on self-identification

In bold: p-value considered significant

Table 3.

Univariate analysis of behavioral data of donor candidates and blood donors screened in the study

Variables	Anti-HEV (IgG) positive (%)	Anti-HEV (IgG) negative (%)	ORa (95% CIb)	p-value
Pig consumptionc				0.3735
Yes	9/9 (100)	812/987 (82.27)	4.104 (0.2376 to 70.89)	-
No	0/9 (0)	175/987 (17.73)	-	-
Direct contact with pigc				0.0538
Yes	5/9 (55.56)	251/987 (25.43)	3.665 (1.098 to 12.01)	-
No	4/9 (44.44)	736/987 (74.57)	-	-
Seafood consumptionc				0.7207
Yes	6/9 (66.67)	705/987 (71.43)	0.800 (0.1885 to 2.938)	-
No	3/9 (33.33)	282/987 (28.57)	-	-
Chicken consumptionc				1.00
Yes	9/9 (100)	972/987 (98.48)	0.3028 (0.01686 to 5.440)	-
No	0/9 (0)	15/987 (1.52)	-	-
Sexual orientationd				0.815
Heterosexual	9/9 (100)	944/987 (95.64)	-	-
Homosexual	0/9 (0)	24/987 (2.43)	-	-
Bisexual	0/9 (0)	19/987 (1.93)	-	-
International tripc				1.00
Yes	1/9 (11.11)	138/987 (13.98)	0.7690 (0.0687 to 4.994)	-
No	8/9 (88.89)	849/987 (86.02)	-	-
Lake bathc				0.0258
Yes	5/9 (55.56)	209/987 (21.18)	4.553 (1.391 to 15.25)	-
No	4/9 (44.44)	778/987 (78.82)	-	-
Weir bathc				1.00
Yes	2/9 (22.22)	275/987 (27.86)	0.7397 (0.1544 to 3.308)	-
No	7/9 (77.78)	712/987 (72.14)	-	-

^aOR: odds ratio

^bCI: confidence interval

^cCalculated by the Fisher’s exact test

^dCalculated by the χ² test

In bold: p-value considered significant

Although no association was observed between the other characteristics and HEV serology, we observed that most positive individuals were black (55.56%, 5/9), single (55.56%, 5/9), with high school (44.45%, 4/9), had piped water (100%, 9/9) and artesian well water (66.67%, 6/9), ate pig (100%, 9/9), ate chicken (100%, 9/9), ate seafood (66.67%, 6/9), had direct contact with pig (55.56%, 5/9), were heterosexual (100%, 9/9), did not travel outside Brazil (88.89%, 8/9), and did not bathe in weir (77.78%, 7/9) (Tables 2 and 3).

Discussion

The transmission of HEV to humans occurs mainly through consumption of contaminated food or water [5, 6]. However, secondary transmission routes, such as blood transfusion, have also been described, specifically for viral genotypes 3 and 4 [6, 7, 8]. This is an important issue as blood transfusion recipients are often immunosuppressed and have a higher risk of severe HEV disease [9, 11]. Despite this, studies on HEV in blood donors in Brazil are rare.

Herein, we found a 0.9% seroprevalence for anti-HEV IG. Although it was not possible to investigate the presence of HEV RNA in our study, only HEV genotype 3 has been identified in Brazil so far [24]. This genotype, as mentioned above, has been associated with HEV transmission via blood transfusion. Interestingly, the prevalence of anti-HEV IgG reported in this study is considerably lower than that observed in blood donors from several countries that have genotype 3 as the main HEV genotype [12], such as the USA (7.3%) [17], Upper Austria (13.55%) [15], and France (56.1%) [13].

In addition, the seroprevalence found here was also lower than those described for blood donors from other Brazilian regions: Southeast (4–11%) [25, 26, 27, 28] and Southern region (2.3–40.25%) [22, 29, 30, 31]. The HEV seroprevalence found in our study was also lower than that observed in the only study carried out in Northeastern Brazil, Bahia state (2%) [23]. Only the Northern Brazil had a lower HEV seroprevalence than that reported in this study (0.44%) [21]. In general, these differences may be related to the serological assays, year of study, sampling, and cultural and infrastructure differences among the analyzed areas, especially when considering the continental size of Brazil.

We found a positive association between male sex and HEV seropositivity. This association has also been described in previous studies [13, 14] and may be related to higher behavioral risk factors, such as such as traveling or working outside at home, differences in health-seeking behaviors [32], and/or high meat consumption, which is perceived as more masculine [33]. On the other hand, Fu et al. [34] recently reported that women had a significantly higher HEV seroprevalence than men in blood donors from Yunnan Province, Dali city, China. This different scenario may be related to regional and dietary differences in Dali [34].

We also observed an association between lake bathing and HEV seropositivity. The low prevalence of anti-HEV positive individuals in our study, as well as their distribution between the analyzed categories, makes a more in-depth assessment of this finding difficult, but HEV genotype 3 RNA has been detected in wastewater [35] and an experimental study showed this genotype may remain infectious in manure slurry from concrete pits [36]. An association between income above BRL 20,000/month and HEV seropositivity was also identified in our study population. This result must be carefully analyzed, mainly because this income is about 10 times higher than the average income of Brazilians [37]. Therefore, we encourage future epidemiological studies on HEV to consider the analysis of different income groups, and in higher sampling, in order to verify the consistency of this finding.

Although our study did not find an association between age group and HEV seropositivity, the mean age observed here was very close to that described by previous studies in blood donors from Brazil [25, 26, 27,28,31]. Passos-Castilho et al. [26,31] described an increase in HEV seroprevalence in blood donors from Brazil as age increases and, in general, most cases of HEV genotype 3 infection have been reported in individuals over 50 years of age [38]. We also found no association between behavioral characteristics and serology for HEV. However, most HEV-seropositive individuals claimed to have eaten pig and seafood at some point in their lives and to have had direct contact with pig, practices that coincide with the transmission routes of HEV genotype 3 [6].

Finally, it is important to consider that 77.77% (7/9) of the anti-HEV IgG positive individuals in our study were blood donors. Unfortunately, we cannot define whether the infection occurred before or after their registration as a donor; however, about 43% (3/7) of anti-HEV positive donors made their first donation more than 20 years ago. These data must be taken as a warning sign due to the possibility that these individuals may have been infected after their registration as donors, since naturally occurring infection-induced anti-HEV IgG may persist at detectable levels for about 14 years in only 47–50% of the HEV-infected [39, 40]. The donor identified as “A,”, who made 82 blood donations in the last 23 years (Table 1), illustrates this issue well and corroborates the need for screening for HEV in blood donor candidates.

Conclusion

The seroprevalence found in candidate donors and blood donors from Pernambuco state, Northeast Brazil, was lower than that described for other Brazilian regions, as well as other countries where HEV genotype 3 circulates. Despite this, our finding is a warning sign, mainly because most positive individuals were regular blood donors. Our study showed an association between HEV seropositivity and male sex, and between seropositivity and lake bathing. We also found an association between individual income and HEV seropositivity; to the best of our knowledge, this finding has not yet been described and deserves future epidemiological analysis, especially in larger population groups and with different incomes. Overall, the report of HEV seropositivity, especially for regular blood donors, and the identification of potential risk factors, corroborates the need for screening in Brazilian blood banks. This need is reinforced for blood products intended for immunosuppressed patients, who are at increased risk of severe disease and death from HEV.

Author contribution

Conceptualization: GGC, JVJSJr, and MRCDC; data collection: GGC and ACBM; data analyses: GGC, LAB, JPG, JVJSJr, and MRCDC; serology assay: GGC and LAB; manuscript preparation: GGC, JPG, and JVJSJr; manuscript review: GGC, LAB, JVJSJr, JPG, ACBM, and MRCDC. Supervision: MRCDC. All authors have read and approved the final version of the manuscript.

Funding

GGC and LAB received scholarships from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) (Brazil) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (Brazil), respectively. MRCDC was supported by CNPq research fellowship.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and patient consent

The study was approved by the Ethics and Research Committee of HEMOPE, protocol 4.636.584, and Certificate of Presentation of Ethical Appreciation 40987620.0.3001.5195.

Consent for publication

All authors approved the version to be published.

Conflict of interest

The authors declare no competing interests.