Abstract
We determined the susceptibilities of 144 clinical and 49 environmental Aeromonas strains representing 10 different species to 26 antimicrobial agents by the agar dilution method. No single species had a predominantly nonsusceptible phenotype. A multidrug nonsusceptible pattern was observed in three (2.1%) clinical strains and two (4.0%) strains recovered from diseased fish. Common clinical strains were more resistant than the corresponding environmental isolates, suggesting that resistance mechanisms may be acquired by environmental strains from clinical strains.
TEXT
Aeromonas species are globally distributed Gram-negative, oxidase-positive fermentative rods, found in aquatic environments (15), foods (12), and the microflora of fish (16). Antimicrobial resistance in these organisms is usually chromosomally mediated, but β-lactamases produced by aeromonads may occasionally be encoded by plasmids (11, 22) or integrons (4). These enzymes have activity against most β-lactam antimicrobial agents, including cefepime and other extended-spectrum cephalosporins. Antimicrobial susceptibility reporting for Aeromonas generally followed guidelines for the Enterobacteriaceae until the Clinical and Laboratory Standards Institute (CLSI) recently published recommendations (9). The objective of this study was to determine the antimicrobial susceptibility profiles of commonly used agents against a collection of Aeromonas species from clinical, fish, and environmental sources.
Aeromonas spp. used in this study included 144 clinical isolates (comprising 54 wound specimens, 33 blood specimens, 34 stool specimens, and 23 isolates from miscellaneous specimens) and 49 environmental isolates (from water [n = 43], fish [n = 5], and crab meat [n = 1]). Strains were previously identified phenotypically by extensive biochemical testing (3) and their identities confirmed genotypically from their gyrB and rpoD gene sequences (2). Ten Aeromonas spp. were represented: A. aquariorum (58 strains), A. veronii bv. sobria (49 strains), A. hydrophila (39 strains), A. caviae (36 strains), A. jandaei (3 strains), A. media (3 strains), A. salmonicida (2 strains), and one strain each of A. allosaccharophila, A. bestiarum, and A. schubertii.
Antimicrobial susceptibility testing was performed by the agar dilution breakpoint method as described by the CLSI (8). Antimicrobial agents tested included the following: amikacin, amoxicillin, amoxicillin-clavulanate, cephalothin, cefazolin, cefepime, cefoxitin, ceftazidime, ceftriaxone, ciprofloxacin, gentamicin, meropenem, moxifloxacin, nalidixic acid, nitrofurantoin, norfloxacin, piperacillin-tazobactam, tetracycline, ticarcillin-clavulanate, tobramycin, trimethoprim, and trimethoprim-sulfamethoxazole (Table 1). Susceptibility was defined as absence of growth on solid medium containing any of these antimicrobial agents. Presence of growth indicated nonsusceptibility. E-strips containing doxycycline (AB Biodisk, Solna, Sweden), ampicillin, tigecycline, and colistin (bioMérieux, Marcy l'Etoile, France) were used to determine MICs. Interpretative criteria for tigecycline and ampicillin were derived from those described for the Enterobacteriaceae by the Food and Drug Administration (5), and those for doxycycline were derived from those described by the CLSI (9), as outlined in Table 1 of the E-strip package insert. Interpretative criteria for colistin were from Fosse et al. (10) (MIC of <2 μg/ml was considered susceptible). MIC breakpoints used were as follows (in μg/ml): for tigecycline, susceptible (S), ≤2; intermediate (I), 4; resistant (R), ≥8; for doxycycline, S, ≤4; I, 8; R, ≥16; and for ampicillin, S, ≤8; I, 16; R, ≥32. Escherichia coli ATCC 25922 was used as a quality control organism for both E-strip MICs and agar dilution tests. Statistical analyses were conducted with Fisher's exact method of contingency table analysis using statistical software (Prism version 5.0; GraphPad, Inc., San Diego, CA).
Table 1.
Antimicrobial susceptibilities of 193 Aeromonas species
| Antimicrobial agent | MIC breakpoint(s) (μg/ml) | Percentage (no.) of strains susceptible |
||
|---|---|---|---|---|
| All isolates (n = 193) | Clinical isolates (n = 144) | Environmental isolates (n = 49) | ||
| Amoxicillin (AMX) | 8 | 1.6 (3) | 0.7 (1) | 4.0 (2) |
| Amoxicillin-clavulanate (AMC) | 8/4 | 16.5 (32) | 6.25 (9) | 46.9 (23) |
| Norfloxacin (NOR) | 4 | 100 | 100 | 100 |
| Ciprofloxacin (CIP | 1 | 100 | 100 | 100 |
| Nitrofurantoin (NIT) | 32 | 99.5 (192) | 99.3 (143) | 100 |
| Trimethoprim (TMP) | 8 | 92.7 (179) | 91 (131) | 97.9 (48) |
| Cephalothin (CEF) | 8 | 27.4 (53) | 20.8 (30) | 46.9 (23) |
| Meropenem (MEM) | 0.25 | 100 | 100 | 100 |
| 1 | 100 | 100 | 100 | |
| 4 | 100 | 100 | 100 | |
| Gentamicin (GEN) | 4 | 99.5 (192) | 99.3 (143) | 100 |
| Tobramycin (TOB) | 4 | 95.3 (184) | 93.8 (135) | 100 |
| Amikacin (AMK) | 16 | 100 | 100 | 100 |
| Ceftriaxone (CRO) | 1 | 96.9 (187) | 95.8 (138) | 100 |
| Ceftazidime (CAZ) | 0.5 | 97.4 (188) | 96.5 (139) | 100 |
| 4 | 99.5 (192) | 99.3 (143) | 100 | |
| Aztreonam (ATM) | 4 | 99.5 (192) | 99.3 (143) | 100 |
| Ticarcillin-clavulanate (TIM) | 16/2 | 91.2 (176) | 88.9 (128) | 97.9 (48) |
| 64/2 | 95.9 (185) | 95.1 (137) | 97.9 (48) | |
| Trimethoprim-sulfamethoxazole (SXT) | 2/38 | 98.9 (191) | 98.6 (142) | 100 |
| Cefepime (FEP) | 0.5 | 98.9 (191) | 98.6 (142) | 100 |
| 8 | 100 | 100 | 100 | |
| Nalidixic acid (NAL) | 16 | 96.9 (187) | 97.9 (141) | 93.8 (46) |
| Cefoxitin (FOX) | 8 | 65.2 (126) | 65.9 (95) | 63.2 (31) |
| Piperacillin-tazobactam (TZP) | 16/4 | 97.4 (188) | 96.5 (139) | 100 |
| 64/4 | 98.9 (191) | 98.6 (142) | 100 | |
| Moxifloxacin (MXF) | 1 | 98.9 (191) | 99.3 (143) | 97.9 (48) |
| Tetracycline (TET) | 4 | 94.3 (182) | 95.1 (137) | 81.6 (40) |
| Cefazolin (CFZ) | 2 | 20.7 (40) | 8.2 (9)a | 10.2 (5) |
| Doxycycline (DOX) | S, ≤4; I, 8; R, ≥16 | 97.9 (189) | 97.2 (140) | 100 |
| Tigecycline (TGC) | S, ≤2; I, 4; R, ≥8 | 100 | 100 | 100 |
| Colistin (CST) | S, <2 | 44.5 (86) | 39.5 (57) | 59.1 (29) |
One hundred nine strains tested.
All isolates were inhibited by amikacin, cefepime (8 μg/ml), ciprofloxacin, meropenem, norfloxacin, and tigecycline. Susceptibility to amoxicillin was demonstrated in three (1.6%) isolates (one clinical and one environmental A. veronii bv. sobria and one environmental A. aquariorum isolate) by agar dilution and confirmed by the E-strip method, with MIC values of 8 μg/ml for all three isolates. Thirty-two isolates (16.5%) failed to grow in the presence of amoxicillin-clavulanate, while 17 (8.8%) were nonsusceptible to ticarcillin-clavulanate (16/2 μg/ml). Of these, 8 (4.4%) were also nonsusceptible to the higher concentration of ticarcillin-clavulanate (64/2 μg/ml). Susceptibility to cephalothin and cefazolin was observed in 53 (27.4%) and 40 (20.7%) isolates, respectively. A moderate level of susceptibility was detected with cefoxitin (126 isolates, 65.2%) and colistin (86 isolates, 44.5%). The majority of the isolates were susceptible to the remaining antimicrobial agents (Table 1). The MICs for doxycycline ranged from 0.064 to 24.0 μg/ml, those for tigecycline ranged from 0.064 to 3.0 μg/ml, and those for colistin ranged from 0.094 to >256 μg/ml. Susceptibility to doxycycline and tigecycline was high in clinical strains, at 97.2 and 100%, respectively. There was no statistically significant difference in antimicrobial susceptibility between clinical and environmental isolates of A. aquariorum. In contrast, clinical isolates of A. veronii bv. sobria were less susceptible than environmental strains (P = 0.0226). Other statistically significant differences were observed for amoxicillin-clavulanate between A. aquariorum and A. hydrophila (P = 0.0036) (A. aquariorum was less susceptible than A. hydrophila) and between A. aquariorum and A. veronii bv. sobria (P = 0.0053) (A. veronii bv. sobria was less susceptible than A. aquariorum) but not between A. aquariorum and A. caviae. Further, susceptibility to cephalothin was significantly higher in A. veronii bv. sobria than in A. aquariorum, A. caviae, and A. hydrophila (P = 0.0001). Nine clinical isolates (6.2%) were able to grow in agar plates containing 4 μg/ml of tobramycin, including seven (14.2%) A. veronii bv. sobria, one (2.7%) A. caviae, and one (33.3%) A. media isolate. Multidrug nonsusceptible patterns were observed in three isolates. Of these, A. caviae strain 138 was less susceptible to most β-lactams, including aztreonam. A. veronii bv. sobria strain 189 was the only isolate to grow in the presence of both gentamicin and tobramycin. Susceptibility to colistin was recorded in 57 (39.5%) clinical and 29 (59.1%) environmental isolates. A. caviae was the most susceptible species (83.7%), next to A. aquariorum (31.0%). Most environmental isolates were susceptible to tetracycline (81.6%) and nalidixic acid (93.8%). Moderate susceptibility was observed with amoxicillin-clavulanate (46.9%), cephalothin (46.9%), and cefoxitin (63.2%), while only five (10.2%) isolates were susceptible to cefazolin.
Differences in antimicrobial susceptibility between clinical and environmental strains have been described previously (19, 20). The resistance observed in environmental aeromonads has been associated with heavily polluted waters as the source of multiple resistance plasmids (13). In contrast, our results suggest that (i) environmental strains are not the principal source of resistance but that antibiotic resistance in clinical isolates may be due to the selective pressure to which these organisms may have been exposed, (ii) water sources are less polluted in Western Australia than other regions, and (iii) environmental strains may have acquired resistance determinants from clinical strains.
In general, growth of Aeromonas was inhibited by most antimicrobial agents, with few isolates showing a multidrug nonsusceptible profile. Susceptibility to tetracycline was high (94.36%), consistent with previous reports from Australia and the United States (18, 20). In contrast, tetracycline resistance in up to 49% of isolates has been reported in studies from the Asian region (7, 17, 19). The three amoxicillin-susceptible isolates described here confirm that amoxicillin-susceptible strains other than A. trota (6) occur, as previously reported (1, 14), and that their growth may be suppressed by amoxicillin-containing medium.
Susceptibility to cephalothin was high in A. veronii bv. sobria, a feature that has been reported by others and proposed as a phenotypic marker to differentiate this species from other aeromonads (18, 20). Similarly, susceptibility to colistin was proposed as an identifying marker for Aeromonas (10). Our results were consistent with those obtained by a previous study (10) for A. hydrophila (61.7% resistance in this study, versus 85.8%) and A. jandaei (100% resistance in both studies). However, MIC results obtained in this report differed from the previous study for A. veronii bv. sobria (61.7% versus 2.5%) and for A. caviae (16.2% versus 2.1%). The numbers of isolates susceptible to piperacillin-tazobactam (97.4% and 98.9%) and ticarcillin-clavulanate (91.2% and 95.9%) were much higher than those susceptible to amoxicillin-clavulanate (16.5%), suggesting that the former two antimicrobials could be considered for the treatment of infections caused by Aeromonas. Zemelman et al. (24) reported that, depending on the strain, the MIC to amoxicillin decreased from 2- to 8-fold in combination with clavulanate, thus increasing the activity of this agent. However, prolonged use of amoxicillin-clavulanate to treat infections caused by A. veronii bv. sobria has resulted in overexpression of carbapenemases and cephalosporinases (23).
All isolates were susceptible to meropenem. A single A. hydrophila isolate that grew in all three agar dilution concentrations was susceptible by the E-strip method using two different inocula, 1.5 × 108 CFU/ml and 3.0 × 108 CFU/ml (results not shown). A large inoculum (3 × 108 CFU/ml) has been recommended to detect carbapenemase production before antibiotic therapy using carbapenems is considered, as conventional in vitro susceptibility testing may fail to detect the presence of carbapenemases in otherwise carbapenemase-susceptible phenotypes (21).
In conclusion, this study shows that the number of multidrug nonsusceptible Aeromonas species in Western Australia remains low and that clinicians have a wide choice of antimicrobial agents to treat infections with these species, consistent with other reports (17, 25). However, antimicrobial susceptibility testing for clinically significant strains is highly recommended, as resistance to antibacterial agents may be strain dependent.
Footnotes
Published ahead of print 28 November 2011
REFERENCES
- 1. Abbott SL, Cheung WKW, Janda JM. 2003. The genus Aeromonas: biochemical characteristics, atypical reactions, and phenotypic identification schemes. J. Clin. Microbiol. 41:2348–2357 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Aravena-Román M, Harnett GB, Riley TV, Inglis TTJ, Chang BJ. 2011. Aeromonas aquariorum is widely distributed in clinical and environmental specimens and can be misidentified as Aeromonas hydrophila. J. Clin. Microbiol. 49:3006–3008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Aravena-Román M, Riley TV, Inglis TJJ, Chang BJ. 2011. Phenotypic characteristics of human clinical and environmental Aeromonas in Western Australia. Pathology 43:350–356 [DOI] [PubMed] [Google Scholar]
- 4. Barlow R, Gobius K. 2009. Environmental reservoirs of integrons: the contribution of production environments to the presence of integrons in beef cattle, poster P20-01, p 139 Annu. Meet. Aust. Soc. Microbiol., Perth, Western Australia [Google Scholar]
- 5. bioMérieux April 2010. E-test antimicrobial susceptibility testing for in-vitro diagnostic use. bioMérieux, Marcy-l'Etoile, France [Google Scholar]
- 6. Carnahan AM, et al. 1991. Aeromonas trota sp. nov., an ampicillin-susceptible species isolated from clinical specimens. J. Clin. Microbiol. 29:1206–1210 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Chang BJ, Bolton SM. 1987. Plasmids and resistance to antimicrobial agents in Aeromonas sobria and Aeromonas hydrophila clinical isolates. Antimicrob. Agents Chemother. 31:1281–1282 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Clinical and Laboratory Standards Institute 2006. Methods for the antimicrobial dilution and disk susceptibility testing of infrequently isolated or fastidious bacteria, 2nd ed, M45-A2, vol 30, no 18 Clinical and Laboratory Standards Institute, Wayne, PA: [DOI] [PubMed] [Google Scholar]
- 9. Clinical and Laboratory Standards Institute 2011. Performance standards for antimicrobial susceptibility testing, 21st informational supplement. CLSI M100-S21 Clinical and Laboratory Standards Institute, Wayne, PA [Google Scholar]
- 10. Fosse T, Giraud-Morin C, Madinier I. 2003. Induced colistin resistance as an identifying marker for Aeromonas phenospecies groups. Lett. Appl. Microbiol. 36:25–29 [DOI] [PubMed] [Google Scholar]
- 11. Fosse T, et al. 2004. Aeromonas hydrophila with plasmid-borne class A extended-spectrum beta-lactamase TEM-24 and three chromosomal class B, C, and D beta-lactamases, isolated from a patient with necrotizing fasciitis. Antimicrob. Agents Chemother. 48:2342–2344 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Hänninen ML, Siitonen A. 1995. Distribution of Aeromonas hydrophila phenospecies and genospecies among strains isolated from water, foods or from human clinical samples. Epidemiol. Infect. 115:39–50 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Huddleston JR, Zak JC, Jeter RM. 2006. Antimicrobial susceptibilities of Aeromonas spp. isolated from environmental sources. Appl. Environ. Microbiol. 72:7036–7042 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Huddleston JR, Zak JC, Jeter RM. 2007. Sampling bias created by ampicillin in isolation media for Aeromonas. Can. J. Microbiol. 53:39–44 [DOI] [PubMed] [Google Scholar]
- 15. Khardori N, Fainstein V. 1988. Aeromonas and Plesiomonas as etiological agents. Annu. Rev. Microbiol. 42:395–419 [DOI] [PubMed] [Google Scholar]
- 16. Kirov SM. 1993. The public health significance of Aeromonas spp. in foods. Int. J. Food Microbiol. 20:179–198 [DOI] [PubMed] [Google Scholar]
- 17. Ko WC, et al. 1996. Increasing antibiotic resistance in clinical isolates of Aeromonas strains in Taiwan. Antimicrob. Agents Chemother. 40:1260–1262 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Koehler JM, Ashdown LR. 1993. In vitro susceptibilities of tropical strains of Aeromonas species from Queensland, Australia, to 22 antimicrobial agents. Antimicrob. Agents Chemother. 37:905–907 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. McNicol LA, et al. 1980. Isolation of drug-resistant Aeromonas hydrophila from aquatic environments. Antimicrob. Agents Chemother. 17:477–483 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Motyl MR, McKinley G, Janda JM. 1985. In vitro susceptibilities of Aeromonas hydrophila, Aeromonas sobria, and Aeromonas caviae to 22 antimicrobial agents. Antimicrob. Agents Chemother. 28:151–153 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Rossolini GM, Walsh T, Amicosante G. 1996. The Aeromonas metallo-B-lactamases: genetics, enzymology, and contribution to drug resistance. Microb. Drug Resist. 2:245–252 [DOI] [PubMed] [Google Scholar]
- 22. Sanchez-Cespedes J, et al. 2008. Plasmid-mediated QnrS2 determinant from a clinical Aeromonas veronii isolate. Antimicrob. Agents Chemother. 52:2990–2991 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Sanchez-Cespedes J, et al. 2009. Development of imipenem resistance in an Aeromonas veronii biovar sobria clinical isolate recovered from a patient with cholangitis. J. Med. Microbiol. 58:451–455 [DOI] [PubMed] [Google Scholar]
- 24. Zemelman R, et al. 1984. Resistance of Aeromonas hydrophila to B-lactams antibiotics. J. Antimicrob. Chemother. 14:575–579 [DOI] [PubMed] [Google Scholar]
- 25. Zhiyong Z, Xiaoju L, Yanyu G. 2002. Aeromonas hydrophila infection: clinical aspects and therapeutic options. Rev. Med. Microbiol. 13:151–162 [Google Scholar]