Introduction

Hepatitis E virus (HEV) infections occur worldwide1. In tropical developing countries human-associated HEV genotypes 1 and 2 (HEV-1/2) of the genus Paslahepevirus are mostly transmitted through contact with contaminated drinking water leading to epidemic outbreaks. In contrast, genotypes 3 and 4 (HEV-3/4) of the same genus are mostly transmitted zoonotically in industrialized countries mainly from wild boar and pig to humans through ingestion of undercooked meat1. Furthermore, rat-derived HEV strains (HEV-C1) from the genus Rocahepevirus are of particular importance2 as cases of human infections caused by this variant have been diagnosed in Hong Kong and Spain3. Although numerous studies highlighted the role of the natural reservoir hosts especially pig and wild boar but also deer and rabbits on zoonotic transmission of HEV, significantly fewer studies have investigated the relevance of pets as possible hosts and potential sources of infection (Table 1). In general, numerous viruses that can infect companion animals are also infectious to humans4. This raises the question of the significance of HEV infections in companion animals. An in silico analysis of host genetics and HEV genetics identified dogs, and rats as potentially susceptible to Paslahepevirus infections, while cats and dogs were described as susceptible to HEV-C1 infections5. To get more information on HEV infections in German cats, dogs, and horses, a molecular and serological analysis of serum samples provided by veterinary diagnostic laboratories was conducted.

Table 1 Previous studies concerning the relevance of HEV and anti-HEV positivity in dogs, cats, and horses (chronological order).
Full size table

Material and methods

Sampling

All samples were obtained from a routine veterinary laboratory (Synlab, Berlin, Germany). All samples have been collected in April or May 2022. Only serum samples from which sufficient material for PCR and serology (approx. 1 ml) was available after routine diagnostic could be tested. Initially, the goal was to test 120 dogs, 120 cats and 120 horses, but due to availability of samples there were small variations (dogs n = 124, cats n = 119, horses n = 122). Initially, we had tried to achieve an even distribution between animals with elevated liver values and those without elevated liver values in the first 200 or so animals, but since this was not realistically achieved, we switched to unselected serum samples regardless of how high the liver values were.

Basic characteristics such as sex, age, breed, or serum concentrations of enzymes aspartate transaminase (AST) and alanine aminotransferase (ALT) were provided in the far majority of studied pets. Initially, it was tried to include equal numbers of samples with elevated and normal liver values, but this was not successful so we switched to "unselected" study subjects.

No experiments on vertebrates were performed as part of this study. Only retrospective serum samples obtained for diagnostic purposes were retrospectively analyzed.

Therefore, the ARRIVE guidelines do not apply to the study.

The datasets generated and/or analyzed during the current study are not publicly available, because the data were examined completely anonymously and public access would allow individual authors of this manuscript to identify the animals and their serological status. However, the data are available from the corresponding author on reasonable request.

Due to anonymized testing, none of the authors of this paper can currently assign the serological results to individual animals with names and owners, and this anonymity should be preserved as far as possible.

Completeness of the dataset

There were 2555 data collected for the entire study cohort, of which 484 (19%) were missing. In detail, the following were present: anti-HEV IgG value at 100%, OD value at 100%, AST at 90%, ALT at 85%, sex 68%, age 63%, race 62%.

Serology

Serological testing has been performed by the MP Diagnostics HEV ELISA 4.0 (MP Biomedicals Germany GmbH, Eschwege), a commercial test able to detect anti-HEV in mammals.

Since the serum samples were completely anonymized, the personal data of the animal owners were not known during the testing. Therefore, no written consent of the animal owners is available. However, this is also not necessary. According to German laws, there is an obligation to obtain consent and an ethical vote or animal ethical vote for a prospective study and retrospective analyses, but not for retrospective anonymized studies. After consultation with the Ethics Committee of the Hamburg Medical Association, a formal ethics vote is not required for retrospective, anonymized serum analyses.

Molecular biology

All animals have been tested by the RealStar® HEV RT-PCR Kit 2.0 (Altona Diagnostics, Hamburg Germany). Serological positive serum samples have been re-tested by a second independent SYBR Green-based nested in-house broad range reverse-transcriptase (RT-) PCR, which targets a highly conserved region of the RNA-dependent RNA Polymerase within the ORF1 covering all genera of the subfamily Hepevirinae including genus Rocahepevirus6.

Statistical analysis

Continuous variables with a non-normal distribution were expressed as median and interquartile range (IQR). Groups were compared using the Mann–Whitney U test. Categorical variables were expressed as a number (%) and compared with Fisher’s exact test. p values less than 0.05 were considered statistically significant. Statistical analyses were performed using SPSS, version 21.0 (IBM Corp., Armonk, NY, USA).

Results

In total 365 serum samples of companion animals have been studied. In 246/365 (67%) the sex was known and 55% of these were male (n = 135). In 230/365 the age was known (63%), in 328/365 the AST was known (90%) and in 309/365 the ALT was known (85%) Characteristics of anti-HEV positive and negative animals are depicted in Table 2.

Table 2 Comparison of anti-HEV positive and negative animals*.
Full size table

The anti-HEV seroprevalence determined by the MP assay varied largely between 10% in dogs (12/124), 6% in cats (7/119), and 2% in horses (2/122) (Fig. 1). The difference between seroprevalence rates in dogs vs. horses (p = 0.01) reached statistical significance (Chi-square test).

Figure 1
figure 1

Results of serological and PCR testing.

Full size image

Anti-HEV ELISA OD values were significantly higher in dogs in comparison to cats (p = 0.008) or horses (p < 0.001) and in cats compared to horses (p = 0.008, C, Mann–Whitney test) (Fig. 2). None of the animal serum samples tested PCR positive. All serologic-positive animals were re-tested with a broad range of RT-PCR covering tall genera of the Orthohepevirinae family but did not uncover any positive result.

Figure 2
figure 2

Levels of MP-assay OD values as a surrogate for anti-HEV levels.

Full size image

An analysis of the samples by zip code of origin did not indicate a regional clustering of positive samples. The area of submission covers the whole of Germany and the positive samples came from Baden-Württemberg (n = 2), Bavaria (n = 4), Berlin (n = 2), Brandenburg (n = 2), Hessen (n = 2), Lower Saxony (n = 3), North Rhine-Westphalia (n = 3), Schleswig–Holstein (n = 2).

Discussion

While the relevance of pork consumption for transmission of HEV genotype 3 infections in industrialized nations is well established, the role of domestic animals in the transmission of HEV genotype 3 infections in industrialized nations is still unclear.

The current study demonstrates that the risk of anti-HEV carriage and thus prior exposure to HEV in cats and dogs is between 6 and 10% Similar values of 9.9% for dogs and 2.2% for cats were found in a study from Spain7 using the same commercial ELISA. Such data, based on a veterinary serological assay, provide a clear picture of the relevance of HEV contacts in companion animals, whereas there is a wide variation in results for assays not specifically designed for animals: In dogs, the reported prevalences range from 19% up to 56.6%8 when the most often used Wantai ELISA was applied and 0% up to 23% for in-house ELISA (Table 1). A similar heterogeneous finding is seen in cats when using the Wantai ELISA (3.1%9 over 14.9%10 up to 32.3%10 or in-house ELISA from 2%11 over 11%12 up to 33%13 (Table 1).

The first finding of anti-HEV seropositive horses in Germany highlights the possible infection of equids with HEV. Seroprevalence from other horse studies shows a high variation from 0.4% up to 18.18% (Table 1). The lower seroprevalence in horses compared to dogs and cats may be explained by the fact that dogs and cats live directly in the same household as their owners and are carnivores, making transmission from the owner's meat products conceivable.

A so far underestimated risk could be the increasingly popular BARF diet (“Biologically Appropriate Raw Food”), i.e. feeding raw meat to dogs and cats, which could be a source of infection for parasites, bacteria, and viruses such as Feline and Canine Calicivirus and even then HEV14.

Although this pilot study provides a very good overview of the exposure of dogs, cats, and horses in Germany to HEV, no reliable conclusions can be made concerning individual dog, cat, or horse breeds because this information was missing in 38% of the animals. Prospective larger cohorts are needed if this aspect is to be studied in more detail.

While no significant difference regarding to transaminases could be found in HEV-negative vs. positive animals (Table 2), individual HEV-positive animals display increased liver enzyme values (Table 3). Although these could be interpreted as an expression of ongoing liver inflammation in the context of hepatitis E viremia that has just healed, there are numerous other possible causes. This includes Pancreatitis, Diabetes but also normal age-related changes, that reflect adaptations during the transition from young to adult individuals15,16.

Table 3 Characteristics of individual anti-HEV positive animals (MP-assay).
Full size table

An additional finding from the present study is—similar to all previous reports- that no viral RNA could be detected, which is a strong indication, that these animals can not be considered to be reservoir hosts so far. However, due to their close contact with humans, these animals can be regarded as sentinels, indicating that they share a common source of infection with their owners. Thus risk patients (e.g. transplant recipients) should be informed that dogs and cats—and to a minor degree horses—can indicate and determine risks of HEV exposure for humans.