BACKGROUND: Zoonotic visceral leishmaniasis (ZVL) caused by Leishmania (Leishmania) infantum is an important disease in humans and dogs. Different mammal species are reservoirs but dogs are considered to be the main one. Phlebotomine sand flies are the proven vector. Four systemic insecticides approved for their use in dogs were previously selected based on their potential to be used in endemic countries as part of the control programs of ZVL. These insecticides are proved to be safe and effective against the on-label insects and parasites, but there is no information about their activity against phlebotomine sand flies. METHODS: The phlebotomine mortality of four systemic insecticides in dogs was evaluated using two randomized clinical trials. For the first trial, thirty dogs were randomly allocated into five groups: four treatments and one control, of equal size. The treatments evaluated were: Guardian(R)SR, Elanco (moxidectin); Comfortis(R), Elanco (spinosad); Bravecto(R), Merck Animal Health (fluralaner); and NexGard(R), Merial (afoxolaner). Blood from dogs was taken at days 2, 4, 21 and 31 post-treatment (trial 1). The compound that showed the highest efficacy was selected for a second trial (trial 2) with 20 dogs sampled at days 0, 2, 4, 7, 14, 18, 32, 39, 51 and 84 post-treatment. Membrane feeding bioassays with Phlebotomus papatasi were used to evaluate the phlebotomine mortality efficacy of the different treatments. Phlebotomine mortality was observed every 24 h following the membrane feeding during 5 days. A mixed model for a negative binomial logistic regression, and a Cox proportional hazard mixed model were used to estimate phlebotomine mortality due to different treatments. RESULTS: Fluralaner was the only compound that showed significant phlebotomine mortality. Fluralaner maintained the phlebotomine mortality between 60-80% for 30 days after treatment. In trial 1 we found that fluralaner increased the risk of death by 1.9 times (95% CI: 1.02-3.6) and 1.7 times (95% CI: 1.09-2.6) at days 2 and 4 after treatment. The Cox model resulted in an increase of 1.47 (95% CI: 1.1-1.96) times in hazard risk at day 2 and 1.89 (95% CI: 1.35-2.45) at day 4 after treatment. In trial 2 we found that fluralaner increased the risk of death by 1.64 times (95% CI: 1.16-2.54) and 1.97 times (95% CI: 1.23-3.17) at days 14 and 32. The hazard risk was also increased by 1.92 (95% CI: 1.4-2.64) times at day 14 after treatment. Phlebotomine survival including all experimental days was significantly lower in the fluralaner group in both trials. CONCLUSIONS: A single oral treatment of fluralaner in dogs induces phlebotomine mortality. Systemic insecticides in dogs should be considered as a potential preventive measure of ZVL.