Wild animal mortality monitoring and human Ebola outbreaks, Gabon and Republic of Congo, 2001-2003

Abstract

All human Ebola virus outbreaks during 2001-2003 in the forest zone between Gabon and Republic of Congo resulted from handling infected wild animal carcasses. After the first outbreak, we created an Animal Mortality Monitoring Network in collaboration with the Gabonese and Congolese Ministries of Forestry and Environment and wildlife organizations (Wildlife Conservation Society and Programme de Conservation et Utilisation Rationnelle des Ecosystemes Forestiers en Afrique Centrale) to predict and possibly prevent human Ebola outbreaks. Since August 2001, 98 wild animal carcasses have been recovered by the network, including 65 great apes. Analysis of 21 carcasses found that 10 gorillas, 3 chimpanzees, and 1 duiker tested positive for Ebola virus. Wild animal outbreaks began before each of the 5 human Ebola outbreaks. Twice we alerted the health authorities to an imminent risk for human outbreaks, weeks before they occurred.

Figure 1

Map of the forest zone straddling the border between Gabon and Republic of Congo, showing (red points) the location of Ebola virus–positive carcasses, confirmed by testing in the Centre International de Recherches Médicales de Franceville biosafety level 4 unit during the 2001–2003 outbreaks in Gabon and Republic of Congo.

Figure 2

Field watertight clothes equipped with air filtration equipment, used for high-risk wild animal necropsy. Odzala National Park Republic of Congo, June 2003. Photo: P. Rouquet.

Figure 3

Species distribution of carcasses found in the forest straddling the border between Gabon and Republic of Congo (2001–2003). * = other primates: Cercopithecus sp.; †= other species: Atherurus africanus (1), Genetta sp. (3), Loxodonta africana (1), Manis sp. (1), Mongoose sp. (1), Thryonomys swinderianus (2), Tragelaphus sp. (1), Python sebae (2), and bird of prey (1).

Figure 4

Temporal distribution of carcasses found in the forest straddling the border between Gabon and the Republic of Congo (2001–2003). Two peaks of mortality were observed: the first occurred in the Ekata region (Gabon) from November to December 2001 and the second from December 2002 to February 2003 in the Lossi gorilla sanctuary (Republic of Congo).

Figure 5

State of the wild animal carcasses found in the field, Lossi gorilla sanctuary, Republic of Congo, December 2002. Carcasses decompose very rapidly in the equatorial forest. Photo: P. Rouquet. A) Female chimpanzee, 3 days after death. B) Female gorilla, 7 days after death. C) Female gorilla, 21 days after death.

Figure 6

Schematic representation of the Ebola cycle in the equatorial forest and proposed strategy to avoid Ebola virus transmission to humans and its subsequent human-human propagation. Ebola virus replication in the natural host (a). Wild animal infection by the natural host(s) (b), no doubt the main source of infection. Wild animal infection by contact with live or dead wild animals (c). This scenario would play a marginal role. Infection of hunters by manipulation of infected wild animal carcasses or sick animals (d). Three animal species are known to be sensitive to Ebola virus and to act as sources of human outbreaks, gorillas, chimpanzees, and duikers. Person-to-person transmission from hunters to their family and then to hospital workers (e). The wild animal mortality surveillance network can predict and might prevent human outbreaks. Medical surveillance can prevent Ebola virus propagation in the human population.