Abstract
BACKGROUND: Understanding the factors that affect the host-feeding preferences of triatomine bugs is crucial for estimating transmission risks and predicting the effects of control tactics targeting domestic animals. We tested whether Triatoma infestans bugs prefer to feed on dogs vs. chickens and on dogs vs. cats and whether vector density modified host choices and other vital rates under natural conditions. METHODOLOGY: Two host choice experiments were conducted in small caged huts with two rooms between which bugs could move freely. Matched pairs of dog-chicken (six) and dog-cat (three) were assigned randomly to two levels of vector abundance and exposed to starved bugs during three nights. Bloodmeals from 1,160 bugs were tested by a direct enzyme-linked immunosorbent assay. PRINCIPAL FINDINGS: Conditional logistic regression showed that dogs were highly preferred over chickens or cats and that vector density modified host-feeding choices. The relative risk of a bug being blood-engorged increased significantly when it fed only on dog rather than chicken or cat. Bugs achieved higher post-exposure weight at higher vector densities and successive occasions, more so if they fed on a dog rather than on a cat. CONCLUSIONS: Our findings strongly refute the hypothesis that T. infestans prefers to blood-feed on chickens rather than dogs. An increase in dog or cat availability or accessibility will increase the rate of bug feeding on them and exert strong non-linear effects on R(0). When combined with between-dog heterogeneities in exposure, infection, and infectiousness, the strong bug preference for dogs can be exploited to target dogs in general, and even the specific individuals that account for most of the risk, with topical lotions or insecticide-impregnated collars to turn them into baited lethal traps or use them as transmission or infestation sentinels based on their immune response to Trypanosoma cruzi or bug salivary antigens.