Culling vampire bats is not reducing exposures to rabies

A new study by Daniel Streicker of the University of Georgia and colleagues suggests that culling of vampire bats is not only failing to eliminate rabies, but maybe making the situation worse.

Strategies to controlvampire-bat-transmitted rabies in Latin America include vaccination of humans and livestock and reduction of bat populations by culling. However, isolated populations in remote areas make any control attempts difficult, and often vaccination occurs only in response to deaths.

Mass culling of bats was started in the 1970s and relies on application of vampiricide (an anticoagulant paste),either to captured bats, or to bite wounds of cattle. Treated bats, or bats that feed on treated wounds pick up the anticoagulant, return to their coloniesand the vampiricide spreads to other bats during communal grooming, killing them. As young bats tend to be dependent on their mothers for food and maygroom few other
adults, they are probably less affected by vampiricide. Despite widespread use of vampiricide for 40 years, vampire bat rabies still persists throughout Latin America and theoretically, culling hosts can actually increase disease prevalence through resulting increased dispersal. The effectiveness ofvampiricide relies on several untested assumptions, that adult bats areresponsible for rabies transmission, that rabies transmission scales with bat density and that of the benefits of culling outweigh any increased dispersal between colonies.

The new study related bat colony size to cattle density data and bat culling history and then looked at rabies exposure prevalence (measured by testing for rabies neutralizing antibodies) and how that related to individual, colony and local cattle density characteristics.Most colony sizes stayed stable over time, and recapture of bats (always at the same roost) occurred over multiple years. Colony size was highly related to thelocal cattle density, supporting theories that cattle provide resources to support larger colonies. No evidence was found to relate colony size to culling history.

Antibodies against rabies (i.e.evidence of exposure) were found in 10.2% of 1086 samples analyzed, and all well sampled colonies, even small ones, showed rabies exposures. Exposure rates were generally stable within each colony over the time period studied, with a few exceptions where local viral extinctions and reintroductions could have occurred.

Only two types of data measured helped to explain whether an individual bat had been exposed to rabies: the age of the bat and the culling history of the colony. Rabies exposure was not related to colony size. Interestingly, juvenile and sub-adult bats had evidence of higher rabies exposure than adult bats. Bats from periodically culled colonies actually showed higher rabies exposure levels than those from colonies that were never culled. Taken together, these data suggest that culling could increase the proportion of young bats in a colony, which increases the age class most susceptible to rabies. Regardless of the exact mechanism, there was no evidence to suggest that previous culling attempts had reduced rabies exposures in colonies.

It is possible that culling was simply insufficient to impact colony size, and that increased culling effort could eliminate rabies from vampire bats. However, after 95% of bats were killed using cyanide in an area of Argentina, rabies deaths still occurred less than a kilometer away, suggesting only very localized effects. Data from this study suggests that the culling effort required to be effective may be highly impractical in the mountainous terrain of these study areas. Streicker’s team will continue their monitoring efforts in Peru and is currently developing newepidemiological models that aim to develop more effective strategies for vampire bat rabies control in Latin America.

Summarized by Louise Taylor from ‘Ecological and anthropogenic drivers of rabies exposure in vampire bats: implications for transmission and control’ by Daniel Streicker et al. Proceedings of the Royal Society B (2012) 279, p3384–3392