Prevalence and 9‐year incidence of hepatitis E virus infection among North Italian blood donors: Estimated transfusion risk

Recent European guidelines recommend that screening policiesfor Hepatitis E virus (HEV) in blood donors should be based on local risk assessments. We determined the prevalence of current and past HEV infection in donors from Lombardy, the Italian region providing 24% of the Italian blood supply. We also calculated the incidence of infection over a period of 10 years, and estimated the risk of transfusion-related transmission. The study was conducted within the framework of BOTIA, an EU-funded project. HEV RNA was detected by individual donation testing, and the prevalence and incidence of anti-HEV antibodies were determined in two subgroups. The risk of receiving an infected blood unit was estimated on the basis of HEV RNA yields and serology. RESULTS: One of the 9726 donors was truly viremic. The prevalence of confirmed anti-HEV IgG reactivity was 52/767 (6.8%; 95%CI 5.1-8.8%). The incidence of HEV infection was 7.6/10000 per year (95%CI 2.1-2.5 per year). The estimated transfusion-related risk of infection was 1/10000 blood donations on the basis of HEV RNA yield (upper limit of the 95%CI 1:1666), and 1/16666 donations on the basis of the incidence data (95%CI 1:435-1:57000).In conclusion, The frequency of current and past HEV infection in blood donors living in Northern Italy is among the lowest so far reported in Europe. The estimated transfusion-related risk of infection was similar regardless of whether it was calculated on the basis of HEV RNA yield or serological incidence, thus suggesting stable infection pressure over the last ten years.


| INTRODUC TI ON
Hepatitis E virus (HEV) is mainly spread in humans by contaminated food and water, but it is increasingly being recognized as a threat to blood transfusion safety because of its documented transmission by means of viremic blood components. 1,2 The risk of transfusion-related infection is generally estimated on the basis of the prevalence of HEV RNA among blood donors.
Nucleic acid testing (NAT) has detected a high rate of viremic donations (up to one in 600) in a number of European countries. 1,2 There is some evidence of high prevalence of viremia and anti-HEV reactivity among donors in Abruzzo (Central Italy), 3 although a retrospective analysis conducted in plasma pools by the Italian National Blood Centre suggests that the pattern of HEV circulation might be different in other Italian regions. 4 However, pooling procedures can limit the analytical sensitivity of NAT and so donations should undergo individual testing. Furthermore, a number of reports indicate that the infection pressure is not stable over time. This implies that NAT yields determined in the relatively short time frame of a prevalence study may not be entirely representative of the risk of transfusion-related transmission, and serological incidence data may be more useful.
Taking advantage of a longitudinal biorepository financed by the European Union, 5 we calculated the prevalence and incidence of HEV infection over the last ten years in donors from Northern Italy and used these data to estimate the risk of the transfusion-related transmission of HEV infection.

| SUBJEC TS AND ME THODS
The study was conducted within the framework of the EU-funded year. The risk of receiving an infectious blood unit was estimated using two methods: HEV RNA yield and serological incidence, assuming a viremia duration of four weeks in the case of asymptomatic infections. 6 Figure 1 summarizes the study results.

| RE SULTS
A total of 9726 donor samples were collected for the HEV RNA study: 7253 (74.5%) taken from males and 2473 (25.5%) taken from females, with a mean age of 43 years (range: 18-67).
The ID-NAT assay showed that ten of the samples were initially reactive, but repeated testing confirmed reactivity in only one.
The donor was a 63-year-old female who had normal alanine aminostransferase levels (ie 22U/L) and was fully asymptomatic at the pre-donation examination. She subsequently seroconverted to being anti-HEV IgG positive. None of the nine donors whose initial reactivity was not confirmed seroconverted during the follow-up period.
In addition, 767 samples (76.7% males, 23.3% females, mean age 43 years) were analysed in order to determine the prevalence

| D ISCUSS I ON
Our study combined HEV viremia and serological data in order to estimate transfusion risk over a ten year period. Using high sensitive individual testing, we found that only one of the ten initially HEV RNA reactive samples in our series was found to be truly viremic, accounting for one out of almost ten thousand blood donations. In addition, the overall seroprevalence of anti-HEV IgG Given that nine of the ten initially NAT-reactive samples were not confirmed by further testing or at serological follow-up, it seems that a single determination of HEV RNA has little positive predictive value. The impact of false-positive results might be more evident when testing samples at low risk of infection, like blood donors. It is therefore possible that some of the variability in the prevalence rates recorded in previous studies is related to methodological differences.
As expected, the prevalence of anti-HEV increased with age, thus reflecting a cohort effect and the long-term persistence of antibody reactivity after primary infection. The fact that the age-related increases in prevalence were relatively uniform (data not shown) suggests that no major outbreak has occurred since the 1950s. Our prospective repository of biological samples allowed us to calculate the incidence of HEV infection over a ten year period, which proved to be 7.6/10 000 per year and is substantially lower than the calculated incidence in other European countries such as Germany Lombardy is the most highly populated region in Italy, and its transfusion system provides 24% of the total Italian blood supply. Although the circulation of HEV is minimal in comparison with other European countries, our data indicate that some tens to hundreds of HEV-infected blood components a year may be transfused into blood recipients. Some of them, such as immunocompromised patients, are susceptible to the development of acute or chronic liver failure, or a chronic infection that may rapidly progress to liver cirrhosis and death. 1,2 The findings of this study may be useful for the regional and national blood authorities responsible for making policy decisions given that recent European guidelines 1 recommend that HEV screening policies should be based on local risk assessment studies. Whether or not to introduce HEV NAT screening therefore requires careful consideration: donor screening may very effectively minimize iatrogenic HEV infection, but it is very costly and can be expected to have a relatively minor impact on the number of HEV infections in the population as a whole because the vast majority of new infections seems to be due to dietary exposure. 1,2

CO N FLI C T O F I NTE R E S T
Grifols Italia S.p.A (Milan, Italy) and Diagnostic BioprobesSrl, (Milan, Italy) provided the kits for serological and molecular analyses.