In vitro antimicrobial susceptibility of clinical isolates of Aeromonas caviae, Aeromonas hydrophila and Aeromonas veronii biotype sobria.

Sir, The genus Aeromonas comprises 14 species, although its taxonomy has not yet been resolved. The main species considered to be pathogenic in humans are Aeromonas hydrophila, Aeromonas caviae and Aeromonas veronii biotype sobria. These species can cause both gastrointestinal and extraintestinal infectious diseases. Aeromonas gastroenteritis is generally self-limiting and, except in immunocompromised patients, antibiotic treatment is unnecessary. However, for extraintestinal infections the susceptibility patterns should be known in order to impliment appropriate therapy. The main objective of this study was to evaluate the activity of 24 antimicrobial agents against A. caviae, A. hydrophila and A. veronii biotype sobria clinical isolates. Forty-three Aeromonas spp. clinical isolates were studied. Strains were distributed by species according to sample source as follows: 19 A. caviae strains (14 isolated from faeces, three from blood, one from an abscess, one from cellulitis), 14 A. veronii strains (12 from faeces, one from blood, one from an ulcer) and 10 A. hydrophila strains (seven isolated from faeces, two from joint fluids, one from a wound). All strains were identified to the species level by the 16S rDNA-RFLP method. Dade-microscan Combo Urine IS panels (Dade Behring, West Sacramento, CA, USA) were used for susceptibility testing. The panels were inoculated according to the manufacturer’s guidelines and incubated overnight in the Walk-away System. Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were used as quality control strains. The results of testing the 43 Aeromonas spp. strains against the 24 antimicrobial agents are shown in the Table. Regarding the -lactam antibiotics, the Aeromonas spp. strains analysed were uniformly resistant to ampicillin, as expected. Piperacillin was more active than ticarcillin against all species. The percentage of piperacillin-susceptible strains was 93%, whereas the percentage of ticarcillinsusceptible strains ranged from 14% to 74%, A. caviae being more susceptible than A. veronii and A. hydrophila. The combination of clavulanic acid and amoxicillin enhanced antibacterial activity, whereas tazobactam did not enhance the effect of piperacillin. All the cephalosporins tested except cefazolin showed very good activity against the different Aeromonas spp. tested. Susceptibility to cefazolin was higher for A. veronii (79%) than for A. caviae (53%) and A. hydrophila (40%), in agreement with previously described data. As well as reporting a similar percentage of susceptibility to cefazolin for A. veronii (83%), Overman & Janda also found similar results with Aeromonas trota (80%). All strains tested were susceptible to aztreonam and imipenem. In a previous study analysing 12 clinical isolates of A. veronii, 67% were resistant to imipenem. Among the aminoglycosides, gentamicin and amikacin were more active than tobramycin. This tobramycinresistant, gentamicin-susceptible duality has also been observed in strains isolated in Australia, Taiwan and the USA. In previous studies, fluoroquinolones showed good activity against all species of Aeromonas. We obtained similar results in which all the strains analysed were susceptible to ofloxacin and ciprofloxacin. However, 26% and 20% of the strains of A. caviae and A. hydrophila were resistant to nalidixic acid and pipemidic acid, whereas the resistance was higher in A. veronii strains, with 88% (P 0.05 for the resistance to nalidixic acid of A. veronii compared with the other species) of the strains being resistant to nalidixic acid and pipemidic acid. Resistance in environment-isolated Aeromonas was found in 59% of strains analysed. In summary, although fluoroquinolones have been reported as the first choice treatment for Aeromonas infections, microorganisms resistant to nalidixic acid and susceptible to ciprofloxacin are known to already have a Correspondence Journal of Antimicrobial Chemotherapy (2002) 49, 697–702