Pyrethroid resistance distribution in Triatoma infestans and environmental association along the Argentine endemic...

Insecticide resistance is considered a barrier to chemical control of Triatoma infestans, the main vector of Chagas disease in the Southern Cone of South America. Although initiatives to reduce the incidence of the disease in the region have integrated different strategies, they have mainly relied on vector elimination using pyrethroid insecticides such as deltamethrin. Reports of pyrethroid resistance in connection with T. infestans control failures first emerged in northern Argentina and southern Bolivia. Recently, a mosaic pyrethroid-resistant focus has been described in the center of the Argentine Gran Chaco (Department of General Güemes, Chaco Province), characterized by the presence of susceptible and very highly resistant populations in the same area. The involvement of different resistance mechanisms has been proposed, together with the contribution of environmental variables that promote the toxicological heterogeneity described. In the endemic zone of Argentina, however, new questions arise: Are there any other clusters of resistance? Is there a relationship between the distribution of resistance and environmental variables (as has been observed at smaller scale)? We studied toxicological data from insects collected and analyzed at 224 localities between 2010 and 2020 as part of the resistance monitoring conducted by the Chagas National Program. The sites were classified according to the survival rate of insects exposed to a discriminant dose of deltamethrin: 0-0.19 were considered susceptible, 0.2-0.79 low-resistance, and 0.8-1 high-resistance. Localities were georeferenced to describe the spatial distribution of resistance and to identify environmental variables (demographics, land use, urbanization, connectivity, and climate) potentially associated with resistance. We used Generalized Linear Models (GLMs) to examine the association between resistance and environmental predictors, selecting error distributions based on the response variable definition. For the entire period, 197 susceptible localities were distributed across the endemic zone. Localities with different survival rates were found throughout the area; 9 high-resistance localities circled the two previously identified resistant foci, and 18 low-resistance in 6 provinces, highlighting their relevance for control planning. Precipitation variables were linked to resistance in all the GLMs evaluated. Presence/absence models were the most accurate, with precipitation, distance from the capital city, and land use contributing to the distribution of resistance. This information could be valuable for improving T. infestans control strategies in future scenarios characterized by unpredictable changes in land use and precipitation.