- Top of page
Arenaviruses (genus Arenavirus, family Arenaviridae) and hantaviruses (genus Hantavirus, family Bunyaviridae) are zoonotic viruses that are spread by aerosolised excreta of rodents and, in the case of hantaviruses, also other small mammals. They cause chronic infections with no apparent harm in their natural hosts. Lassa virus, the most prominent arenavirus, causes Lassa fever in West Africa, where up to 300 000 clinical cases are estimated to occur annually (McCormick 1999). Other arenaviruses – such as Lujo, Junin, Machupo, Guanarito and Sabia viruses – are causative agents of haemorrhagic fevers in South Africa and America. The type species of the genus, Lymphocytic choriomeningitis virus, occasionally causes acute central nervous system disease and congenital malformations (Charrel & de Lamballerie 2003; Gunther & Lenz 2004; Paweska et al. 2009). Hantaviruses cause haemorrhagic fever with renal syndrome (HFRS) in Asia and Europe, and hantavirus cardiopulmonary syndrome in the Americas (Schmaljohn & Hjelle 1997; Kruger et al. 2001 Kruger et al. 2011; Peters & Khan 2002).
Recently, several arenaviruses and hantaviruses were found in Guinea, West Africa. Lassa virus was detected in Mastomys natalensis (Lecompte et al. 2006) and a novel arenavirus, Kodoko virus, was found in Mus minutoides (Lecompte et al. 2007). The first indigenous African hantaviruses were discovered in Guinea too; Sangassou virus in Hylomyscus simus (Klempa et al. 2006, 2012) and Tanganya virus in a shrew, Crocidura thereseae (Klempa et al. 2007).
The identification of the first African hantaviruses in 2006/07 raised the question of the relevance of hantaviruses as rodentborne pathogens additional to Lassa virus. To assess public health relevance of these two pathogens, we performed a human seroprevalence study in a subpopulation of recently febrile patients. Very recently, intensive serological search provided evidence of human hantavirus infections and HFRS-like clinical cases in Forest Guinea where Sangassou virus (SANGV) had been detected (Klempa et al. 2010). In the current seroepidemiological study, we focused on another area of Guinea: the savannah (Upper Guinea), a region where Lassa virus and Tanganya virus have been found.
- Top of page
In the Faranah region of Upper Guinea, the human IgG prevalence for Lassa virus is 40%, which is similar to the seroepidemiological data collected 13 years earlier in Gbetaya (43% seroprevalence) and Sangoyah (42% seroprevalence), two villages near those in our study (Lukashevich et al. 1993). The two other villages included in the previous study and showing lower seroprevalences, Tindo (33%) and Kamaraya (25%), are located along the main road but not in deep savannah. Epidemiologically, this seems to have an impact because the villages along the road are mainly infested by the black rat (Rattus rattus) and the domestic mouse (Mus musculus), which are not reservoirs of Lassa virus.
In our study in May 2004, the Lassa IgG seroprevalence values did not significantly differ in the age classes from 20–29 to 70–79 years. This finding suggests that most of the infections occur already during young age (until age 20). A possible decline in antibody prevalence with age might be compensated by few primary and secondary infections in adults, as evidenced by 7 IgM-positive adults in our study. The IgG seroprevalence of 30 – 50% in the savannah region differs mainly from those observed in forest Guinea, where the IgG prevalences were 5 – 30% (ter Meulen et al. 1996; Kerneis et al. 2009).
To determine Lassa fever incidence, we also analysed IgM seropositivity, which indicates recent infection. Lassa virus IgM usually peaks at 10–30 days after onset of fever and then decreases after 80 days (Wulff & Johnson 1979; Johnson et al. 1987). In this study, the Lassa IgM seropositivity reached 2.8%. To our knowledge, this is the first evaluation of Lassa IgM seropositivity of the rural population in a population-based study.
For hantavirus infections, the observed IgG seroprevalence of 1.2% is identical to that found in the forest of Guinea (Klempa et al. 2010), but much lower than that reported earlier from Gabon (8%; Dupont et al. 1987) and Central African Republic (4%; Gonzalez et al. 1988). The latter values probably are overestimates of antibody prevalence because no confirmatory tests were performed. On the one hand, a battery of different serological assays seems to be necessary to avoid false-positive results that might be caused by unspecific reactivity of sera from African people. On the other hand, it needs to be stated that a battery of assays might considerably decrease the sensitivity. However, in this very first assessment in the area, our priority was to clearly confirm whether human hantavirus infections occur. We therefore focused on specificity rather than sensitivity. Nevertheless, a seroprevalence of 1–2% is typical for hantavirus-endemic countries in other parts of the world, such as Germany (Zöller et al. 1995).
TANGV is, at the moment, the only hantavirus known to be present in the investigated area. Whether the hantavirus antibodies originate from infection with TANGV or another so far unknown hantavirus remains unclear. None of the sera exhibited measurable neutralising antibodies against PUUV, HTNV or SANGV (data not shown). Unfortunately, neutralisation tests with TANGV could not be performed as we failed to isolate TGNV on cell culture. These findings indicate that the observed weak reactivity in IFA (but also ELISA and Western blot) is due to the absence of autochthonous antigens in the tests formats and represents cross-reactivity to the heterologous hantavirus antigens used in the assays. The pathogen(s) causative of the human infections were most likely not represented by the panel of used antigens and viruses, and should be novel, more distantly related hantavirus(es).
The reservoir of Lassa virus, M. natalensis, is living in close contact with humans (Fichet-Calvet et al. 2007). This seems not to be the case for hantavirus reservoirs. The reservoir of TANGV, C. thereseae, is living in the savannah, forest and farms, but without invading the houses (Fichet-Calvet et al. 2010). The remarkable difference in seroprevalence rates for Lassa virus and hantaviruses suggests their dissimilar epidemiological characteristics. Most likely, a difference in frequencies for rodent/shrew – human contact could explain this phenomenon as the viruses are transmitted by different reservoir hosts. In agreement with the previous study from Forest Guinea (Klempa et al. 2010), the low hantavirus-specific seroprevalence suggest that these viruses are probably not transmitted (peri)domestically in villages. Other parameters such as differences in virus prevalence in rodents, virus shedding efficiency, virus stability and infectivity might contribute to the contrasting seroprevalence rates of the two viruses, too.
Despite the limitation in terms of the sample size, our study confirms that the South of Upper Guinea is a highly endemic zone for Lassa fever, as previously indicated by the report of four acute cases in Faranah's regional hospital in 1996–1999 (Bausch et al. 2001). This indicates that preventive measures focused on rodent control need to be considered by local public health authorities. Moreover, human hantavirus infections also occur in the region. Their public health relevance still needs to be determined. Further field studies leading to identification and isolation of local hantaviruses will allow to improve diagnostics and to assess their clinical relevance.