Antimicrobial Susceptibility of Flavobacterium psychrophilum from Chilean Salmon Farms and Their Epidemiological...

Flavobacterium psychrophilum is the most important bacterial pathogen for freshwater farmed salmonids in Chile. The aims of this study were to determine the susceptibility to antimicrobials used in fish farming of Chilean isolates and to calculate their epidemiological cut-off (CO_WT) values. A number of 125 Chilean isolates of F. psychrophilum were isolated from reared salmonids presenting clinical symptoms indicative of flavobacteriosis and their identities were confirmed by 16S rRNA polymerase chain reaction. Susceptibility to antibacterials was tested on diluted Mueller-Hinton by using an agar dilution MIC method and a disk diffusion method. The CO_WT values calculated by Normalized Resistance Interpretation (NRI) analysis allow isolates to be categorized either as wild-type fully susceptible (WT) or as manifesting reduced susceptibility (NWT). When MIC data was used, NRI analysis calculated a CO_WT of ≤0.125, ≤2, and ≤0.5 μg mL^-1 for amoxicillin, florfenicol, and oxytetracycline, respectively. For the quinolones, the CO_WT were ≤1, ≤0.5, and ≤0.125 μg mL^-1 for oxolinic acid, flumequine, and enrofloxacin, respectively. The disk diffusion data sets obtained in this work were extremely diverse and were spread over a wide range. For the quinolones there was a close agreement between the frequencies of NWT isolates calculated using MIC and disk data. For oxolinic acid, flumequine, and enrofloxacin the frequencies were 45, 39, and 38% using MIC data, and 42, 41, and 44%, when disk data were used. There was less agreement with the other antimicrobials, because NWT frequencies obtained using MIC and disk data, respectively, were 24 and 10% for amoxicillin, 8 and 2% for florfenicol, and 70 and 64% for oxytetracycline. Considering that the MIC data was more precise than the disk diffusion data, MIC determination would be the preferred method for susceptibility testing for this species and the NWT frequencies derived from the MIC data sets should be considered as the more authoritative. Despite the high frequency of isolates showing full susceptibility to florfenicol, the significant frequencies of isolates exhibiting reduced susceptibility to oxytetracycline and quinolones may result in treatment failures when these agents are used.