Hepatitis E Virus and Person‐to‐Person Transmission

To the Editor—I have read with interest a paper by Teshale et al [1] in a recent issue of the journal. The authors report their findings during a large outbreak of hepatitis E virus infection in northern Uganda and indicate that these findings provide evidence supporting person-to-person transmission of this infection, which has previously been believed to be infrequent [2, 3]. In their study [1], the primary arguments in support of person-to-person transmission of hepatitis E virus are that (1) the secondary attack rate of hepatitis E virus infection within households was high, (2) persons residing in households with multiple cases of hepatitis E virus infection and persons residing in households with a single case or no cases of hepatitis E virus infection had differences in certain exposure or behavioral characteristics, and (3) no continuing source of infection (eg, a source of contaminated water) was found. However, some of these findings may have alternative explanations other than person-to-person transmission. The secondary attack rate of an infection within households, interpreted as the proportion of other members in a household with infection after the occurrence of an index case in that household, depends heavily on the overall disease rate in the population. For instance, if a disease universally affects the members of a population, the secondary attack rate would appear to be 100%, irrespective of the route of transmission of infection. In the hepatitis E outbreak in Uganda under consideration, the overall disease attack rate in the population was 30.5%. The observed secondary attack rate of 25% among household members with an index patient is thus not unusually high and does not by itself suggest person-to-person transmission. The authors also reported that an increasing number of cases in a household was associated with increasing family size [1]. This too cannot be taken as evidence of person-to-person transmission. In any disease outbreak, the number of cases in a household, which, on average, is likely to equal the product of the overall attack rate in the population and the size of a particular household, would be expected to increase with family size, irrespective of the presence or absence of person-to-person transmission within households. Similarly, the authors' findings of differences between persons residing in households with multiple cases of hepatitis E virus infection and persons residing in households with a single case of hepatitis E virus infection may not support the notion of person-to-person transmission. Persons residing in households with multiple cases were found to be more likely to have had a close contact with a jaundiced person, to have attended a funeral of a jaundiced person, to have washed hands in a common basin, or to not have washed hands after defecation, compared with persons residing in households with a single case or no cases. However, the former group lived in larger households than did the latter group. Household size is well known to be associated with socioeconomic status, standard of living, health-related behavior, and the health of its members [4, 5]. Thus, it is quite possible that the differences observed between the 2 groups were related to a confounding effect of household size. In particular, communal basins may exist more often in larger households than in smaller households. Also, the difference between the 2 groups with regard to close contact with a jaundiced person may not be a finding but, instead, may be related to the defining characteristics of the 2 groups (ie, the number of cases in the household). The failure to detect hepatitis E virus RNA in water is not unusual. The concentration of hepatitis E virus in contaminated water is low. Although several studies have reported successful amplification of hepatitis E virus genomic sequences from contaminated water and sewage, the sensitivity and reproducibility of such identification techniques remains unclear. In addition, it may be pointed out that nearly half of the later cases in the Ugandan study [1] occurred from 8 to >20 weeks after the index cases in the same household. Since the maximum incubation period of hepatitis E virus is 10 weeks, several of these delayed cases are unlikely to be related to person-to-person transmission of infection from the index cases in the same household. Person-to-person transmission of hepatitis E virus, a virus with fecal-oral transmission and an oral route of entry, is clearly plausible. The only issue at hand is the frequency of such transmission and the proportion of cases of disease that it accounts for. This proportion may vary under different environmental and socioeconomic conditions. However, as argued here, the findings presented in Teshale et al [1] do not necessarily support the notion that person-to-person transmission played a major role during the outbreak in northern Uganda. Potential conflicts of interest. R.A.: no conflicts.