Abstract
In this cross-sectional study, we investigated veterinarians in Estonia for evidence of exposure to hepatitis E virus (HEV). In 2012, we collected sera from 158 persons attending a veterinary conference, of whom 156 completed a questionnaire covering their background information. Altogether 115 persons reported they had obtained a veterinary degree and were included in this study. The sera were tested for presence of antibodies against HEV using a commercial enzyme linked immunosorbent assay and a commercial immunoblot assay in series. A sample was considered antibody-positive if it tested positive with both tests. Antibody-positive samples were further examined for the presence of HEV RNA. Three (2.6%) of the 115 veterinarians tested positive for immunoglobulin G antibodies against HEV, whereas no immunoglobulin M antibodies against the virus were detected. The antibody-positive veterinarians were small animal practitioners. Pigs comprised no or small part of their working time or patients. No HEV RNA was detected in the antibody-positive samples. The prevalence of antibodies against HEV in veterinarians in Estonia was lower than what has been observed in veterinarians in other countries.
Introduction
G
In the period 2014–2016, 3 of the 14 cases of hepatitis E detected in Estonia were considered imported (National Institute for Health Development, unpublished data). We have detected evidence of exposure to HEV in Estonia in domestic pigs and wild boars, as well as farmers working with pigs and hunters (Ivanova et al. 2015). Moreover, we detected HEV genotype 3 RNA in pig feces and in wild boar samples (Ivanova et al. 2015).
Veterinarians encounter zoonotic pathogens in their work and are often mentioned to be in risk for HEV exposure (Meng et al. 2002; Bouwknegt et al. 2008; Krumbholz et al. 2014; Mesquita et al. 2014; De Sabato et al. 2017; European Centre for Disease Prevention and Control 2017; EFSA Panel on Biological Hazards et al. 2017; Kantala et al. 2017; Lange et al. 2017). Pigs and wild boars are considered the most important reservoirs of HEV, but HEV or anti-HEV antibodies have also been detected in other animals, including pet animals (EFSA Panel on Biological Hazards et al. 2017). In this cross-sectional study, we investigated veterinarians in Estonia for evidence of exposure to HEV.
Materials and Methods
Ethics statement
The study was a part of a larger study on zoonoses, approved by the Research Ethics Committee of the University of Tartu (nr. 216/T-15, 227/M-5, and 235/M-26). Participation was voluntary, and the participants gave written informed consent.
Setting
Estonia is a country in northeastern Europe. Approximately 1000 veterinarians are licenced to work in the country (Estonian Veterinary Board 2016).
Samples
Sera were collected from 158 individuals attending an annual national veterinary conference (Veterinaarmeditsiin) in October 2012 (Lassen et al. 2016). The participants were requested to complete a questionnaire covering their age, gender, living place, education, work, and free time, with special emphasis on plausible risk factors for a range of zoonotic pathogens. In this study, participants who reported in the questionnaire that they have obtained a veterinary degree were defined as veterinarians (n = 115). The other participants were attendees of the veterinary conference who did not explicitly report that they had obtained a veterinary degree and included veterinary students, individuals reporting doing assisting work at veterinary clinics, and individuals who had chosen not to describe their education and work in detail.
Serology
The sera were analyzed for immunoglobulin G (IgG) antibodies against HEV using a commercial sandwich enzyme linked immunosorbent assay (ELISA) (recomWell HEV IgG; Mikrogen GmbH, Neuried, Germany). Sera positive for anti-HEV IgG were also tested for anti-HEV immunoglobulin M (IgM; recomWell HEV IgM; Mikrogen GmbH). ELISA-positive sera were further tested using an immunoblot assay (recomLine HEV IgG and recomLine HEV IgM; Mikrogen GmbH). A sample was defined IgG positive if both the IgG ELISA and IgG immunoblot assay were positive. A sample was defined IgM positive if both the IgM ELISA and IgM immunoblot assay were positive.
Detection of HEV RNA
Samples that were either IgG positive or IgM positive were tested for the presence of HEV RNA. RNA was extracted from 140 μL of serum using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions, and frozen to −80°C. To detect HEV RNA, the qScript™ One-Step Fast qRT-PCR Kit, ROX™ (Quanta BioSciences, MD), and Applied Biosystems 7500 (Applied Biosystems, CA) real-time PCR system were used (Ivanova et al. 2015). Primers and probes were those previously described (Jothikumar et al. 2006).
Statistical analysis
Open-source software OpenEpi (Dean et al. 2016) was used for statistical analyses. Confidence intervals (95% CI) were calculated using Mid-p exact, and differences between prevalences were evaluated using two-by-two tables. Two-tailed p-values (Mid-p exact) are reported.
Results
Sera were collected from 158 persons attending the veterinary conference, of whom 156 completed the questionnaire, and of whom 115 were defined as veterinarians in this study. Among the veterinarians, there were more females (n = 90, 78.3%) than males (n = 25, 21.7%). The mean age of the veterinarians was 37.9 years (median 35, range 25–58 years). Most of the veterinarians lived in Estonia, in the mainland or on the islands. Their living places included both rural and urban areas. Professional experience ranged from 0.5 to 37 years, and small animal practice (representing >80% of the work time or patients) was the predominant (63.5%) work type. None of the veterinarians worked exclusively or more than 50% of their working time or patient flow with pigs but 22 (19.1%) reported that pigs comprised from 1% to 40% of their working time or patients.
Three veterinarians tested positive for anti-HEV IgG antibodies. The apparent prevalence of anti-HEV IgG among the veterinarians was thus 2.6% (3/115; 95% CI 0.67–6.93). The apparent prevalence among all the participants was 1.9% (3/158; 95% CI 0.49–5.08).
None of the samples tested positive for anti-HEV IgM. None of the three anti-HEV IgG-positive samples contained detectable HEV RNA by real-time PCR. The results are summarized in Table 1.
ELISA, enzyme linked immunosorbent assay; HEV, hepatitis E virus; CI, confidence interval (Mid-p exact); IgG, immunoglobulin G; IgM, immunoglobulin M.
The three anti-HEV IgG-positive individuals were all small animal practitioners (≥95% of working time or patients, Table 2). One of them was male and two were females, and two of them were older and one was younger than the median age among all the veterinarians. All three worked in small animal clinics and one also worked in an animal shelter. Their professional experience was from few to almost 30 years. For one of them, pigs comprised ∼2% of the working time or patients. Two of them were cat owners and two of them were dog owners. Two of them reported traveling to subtropics or tropics during the past decade.
No data for all the potential risk factors for all the veterinarians.
CI, confidence interval (Mid-p exact).
Discussion
The apparent prevalence of anti-HEV IgG antibodies was lower in veterinarians in Estonia (2.6%) than in veterinarians in Finland (34/333; 10.2%, p < 0.01) (Kantala et al. 2017), Norway (21/163; 12.9%, p < 0.01) (Lange et al. 2017), Germany (5/22: 22.7%, p < 0.01) (Krumbholz et al. 2012), and Portugal (36/373; 9.7%, p < 0.01) (Mesquita et al. 2014). The result was also lower than that observed in swine veterinarians in the United States (68/295; 23%, p < 0.001) (Meng et al. 2002) and Italy (8/83; 9.64%, p < 0.05) (De Sabato et al. 2017), and in swine veterinarians and swine farmers in Portugal (5/13; 38.5%, p > 0.001) (Teixeira et al. 2017). The prevalence was not statistically different from that in veterinarians not working with swine (n calculated from the reported prevalence 9.2/153; 6%, p = 0.25) nor from that in veterinarians working with swine (n calculated from the reported prevalence 5.4/49; 11%, p = 0.08) in the Netherlands (Bouwknegt et al. 2008). Moreover, the prevalence was lower than that in pig farm workers (9/67; 13.4%, p < 0.01) but similar to that in hunters (6/144; 4.2%, p = 0.53) in Estonia (Ivanova et al. 2015). Not detecting anti-HEV IgM antibodies nor HEV RNA in the samples is in line with the results of other similar studies (Krumbholz et al. 2012; Kantala et al. 2017).
The small number of seropositive individuals did not allow us to identify and evaluate potential risk factors for testing positive for antibodies against HEV using the questionnaire data. Contact with swine is considered a risk factor for HEV for humans (Meng et al. 2002; Olsen et al. 2006; Chaussade et al. 2013; Song et al. 2014; Ivanova et al. 2015; Khuroo et al. 2016; European Centre for Disease Prevention and Control 2017), but other animals may serve as potential reservoirs as well (Meng, 2010; Wedemeyer et al. 2012; EFSA Panel on Biological Hazards et al. 2017; Kantala et al. 2017). All the seropositive individuals were small animal practitioners. In Finland, being a small animal practitioner appeared to be associated with being HEV-IgG seropositive among veterinarians (Kantala et al. 2017). Our results offer support to this (p = 0.06; Table 2).
Our convenience sample represented veterinarians attending a national veterinary conference, and it is possible that veterinarians working with pigs or exposed to other potential risk factors were underrepresented. Nevertheless, our study provided useful baseline data, which adds to the knowledge on HEV in different potential risk groups and serves to increase awareness. The data from this part of the world (this study and Kantala et al. 2017) suggest a potential occupational exposure to HEV when working as a small animal practitioner, but, for example, foodborne exposure cannot be excluded (European Centre for Disease Prevention and Control 2017; EFSA Panel on Biological Hazards et al. 2017).
Footnotes
Acknowledgments
We thank the veterinarians who participated in the study, the organizers of the conference Veterinaarmeditsiin 2012, and our colleagues who contributed to the study. We particularly thank Valentina Tefanova, who passed away in 2015, for her contributions to the study. Moreover, we thank Pille Paats for technical assistance, and nurses Anu Kuusmann and Marge Reiss for taking the blood samples. The work was supported by European Regional Development Fund programme TerVe 3.2.1002.11. (project EKZE-SS), Health Board of Estonia (VIR TerVE 2012; project ZoonRisk), Estonian Research Council (IUT 42-1), and Estonian University of Life Sciences Base Financing (8P160014VLVP).
Author Disclosure Statement
No competing financial interests exist.