The use of epidemiology for evaluating the performance and administration strategies of topical drugs for controlling sea lice in Chile

摘要

Sea lice is considered the most important ectoparasite that affects salt water farmed salmonids around theworld. Its high economic impact has motivated efforts for its control. Caligus rogercresseyi is the sea licespecies of major economic importance for the salmon industry in Chile. Sea lice control in Chile is largelybased on chemotherapeutants administered as immersion treatments, synthetic pyrethroids andorganophosphates being the most used drugs at the present. In recent years, farmers have reporteddecreased performance of pyrethroid-based treatments. Main causes for treatment failures may be lowsensitivity of sea lice to the anti-parasitic drug, failures in the administration of the product, or reinfestationfrom external sources (i.e. neighbouring farms) after the treatment, especially in areas withintensive salmon farming. The general objective of this thesis was to evaluate the performance of topicaldrugs administered as immersion treatments and to evaluate area-level treatment strategies using fielddata under an epidemiological approach which considered the multi-factorial nature of the sea licedynamics at the farm and the farm management area level. The specific objectives of this thesis were: 1)to assess the performance of the synthetic pyrethroids deltamethrin and cypermethrin on different lifestages of C. rogercresseyi (Chapter 2), 2) to explore the spatial and spatio-temporal variation of C.rogercresseyi‘s response to pyrethroid treatments and examine factors related to this variability (Chapter3), and 3) to evaluate the effect of treatment synchronization on adult sea lice levels up to 8 weeks aftertreatment (Chapter 4).In Chapters 2 and 3 we used the sea lice level one week after treatment as the variable of interest, while inChapter 4 we included sea lice levels from week 2 to 8 after the procedure. Pre-treatment sea lice levels,environmental and management factors were included as predictors in the three cases. In Chapter 2 wecompared the treatment performance between three pyrethroid based products available for the Chileanindustry and with a negative control (i.e. non-treated pen). In Chapter 3 we calculated residuals at thetreatment level and farm effects from a multi-level regression model. Farm effects were subjected to a global cluster analysis (Moran‘s I) and to a purely spatial analysis in order to detect clusters of high sealice levels in the study area. Residuals at the treatment level were included in a spatio-temporal analysis.In Chapter 4 we tested the effect of three different measures of treatment synchronization on the adult licelevels up to 8 weeks after treatment.Pens treated with any of the three pyrethroid products had significantly lower mean juvenile, mobileadult, and gravid female sea lice abundance after treatment compared to the untreated pens; however, theeffect on juvenile lice was less than on mobile stages. No significant differences were observed in thenumbers of juvenile, adult male, and non-gravid female when the three products were compared to eachother; however, cypermethrin exhibited a small, yet significantly greater effect on the gravid femalegroup.The global cluster analysis revealed treatment performance was clustered in space, while the scan spatialstatistic found two areas where the post-treatment adult lice level attributed to the farm effect wassignificantly higher than in the rest of the study area. These spatial clusters remained even once weadjusted for environmental and management predictors. In the spatio-temporal analysis, we found threeclusters; however, they become non-significant when predictors were included in the model.The model containing the treatment synchronization variable, expressed as the number of farms within awindow of 1 week and 10 km seaway distance that reported treatment against sea lice, fitted the modelthe best. Higher intensity of treatment synchronization was significantly associated with lower adult licelevels at weeks 5 and 6 after treatment. This relationship appeared to be linear, suggesting that higherlevels of synchronization may result in lower sea lice levels at these weeks.Findings of this thesis suggest that during 2011 and 2012 approved pyrethroids in Chile were moreeffective on adult than juvenile lice. In general, the three products in evaluation showed a similarperformance, but cypermethrin was associated with better results in the gravid female lice group.We also found treatment performance was clustered in space, and two areas of poor treatment performance were located. These spatial clusters remained even once we adjusted for environmental andmanagement predictors, suggesting unknown factors were causing the clustering in these areas. Potentialfactors that may explain these clusters include lower sensitivity of sea lice populations to pyrethroids andproblems with the application of the product; however, further research is needed to clarify this situation.With respect to strategies of treatment at the area level, our results suggest that treatment synchronizationreduced adult lice levels between weeks 5 and 6 after treatment, which could mean a two-week delay inthe rise of sea lice levels after a treatment.