LETTERS no. ISTN3X; 96% identical), com- previous ones. In addition, S. 4. Ward LR, de Sa JD, Rowe B. A phage-typ- prising one of Tn3-like inverted Enteritidis PT6d resistant to ampi- ing scheme for Salmonella enteritidis. Epidemiol Infect 1987;99:291–4. repeats and putative coding regions cillin was recently reported (10). 5. National Committee for Clinical for transposase, resolvase (also called Relationship between RDNC-a in this Laboratory Standards. Performance stan- repressor), and ampicillin resistance. study and PT6d is unknown, and fur- dards for antimicrobial disk susceptibility The resistance gene encodes a TEM-1 ther investigations will be needed. tests; approved standard-7th ed. NCCLS β document M2-A7, Wayne (PA): The type -lactamase. (The sequence has Transfer of an R-plasmid is a com- Committee; 2000. been registered to DDBJ/GenBank/ mon way for bacteria to acquire drug 6. Ridley AM, Punia P, Ward LR, Rowe B, EMBL with accession no. AB103092.) resistance, and it often affects other Threlfall EJ. Plasmid characterization and Conjugative transferability of aspects such as sensitivity of bacterio- pulsed-field electrophoretic analysis demonstrate that ampicillin-resistant strains p981123 between S. Enteritidis strains phages, as described in this study. of Salmonella enteritidis phage type 6a are was examined by using the parental S. Molecular based surveillance for drug derived from Salm. enteritidis phage type 4. Enteritidis RDNC-a R-AS strain as a resistance in S. Enteritidis needs to J Appl Bacteriol 1996;81:613–8. donor, and three independent S. continue. 7. Helmuth R, Stephan R, Bunge C, Hoog B, Steinbeck A, Bulling E. Epidemiology of Enteritidis strains (PT1; PT4; and virulence-associated plasmids and outer PT21) resistant to nalidixic acid (R- Hidemasa Izumiya,* Naomi Nojiri,* membrane protein patterns within seven N) as recipients. p981123 was trans- Yoshiko Hashiwata,† common Salmonella serotypes. Infect ferable between S. Enteritidis strains Kazumichi Tamura,* Jun Terajima,* Immun 1985;48:175–82. 8. Vatopoulos AC, Mainas E, Balis E, -5 -4 and Haruo Watanabe* at frequencies of 10 to 10 , and the Threlfall EJ, Kanelopoulou M, resulting R-AN transconjugant *National Institute of Infectious Diseases, Kalapothalki V, et al. Molecular epidemiol- showed the same lytic pattern of the Tokyo; and †Hiroshima City Institute of ogy of ampicillin-resistant clinical isolates Public Health, Hiroshima, Japan typing phages as RDNC-a. Thus, of Salmonella enteritidis. J Clin Microbiol 1994;32:1322–5. transfer of p981123 could convert the Acknowledgments 9. Kelley WL, Bastia D. Conformational phage types at least from PT1, PT4, changes induced by integration host factor We thank all the municipal and pre- and PT21 to RDNC-a. Pulsed-field at origin gamma of R6K and copy number fectural public health institutes for provid- gel electrophoresis (PFGE) was done control. J Biol Chem 1991;266:15924–37. ing us with Salmonella enterica serovar 10. Eurosurveillance. Upsurge in Salmonella by using XbaI or BlnI as well, and Enteritidis isolates. We also thank Public Enteritidis outbreaks in England and Wales, RDNC-a strains showed a variety of September to November 2002. Health Laboratory Service, United PFGE profiles. These results suggest Eurosurveillance Weekly, vol. 6; 2002. Kingdom, for kindly providing the typing emergence and prevalence of the 50- Available from: URL: http://www.eurosur- phages and the scheme. kb R-plasmid converting phage types veillance.org/ew/2002/021205.asp to RDNC-a in S. Enteritidis in Japan. This work was partly supported by Address for correspondence: Haruo Watanabe, Previous studies reported correla- grants from Ministry of Health, Labor and Department of Bacteriology, National Institute tion between R-plasmids and phage Welfare, and Ministry of Education, of Infectious Diseases, Toyama 1-23-1, types of S. Enteritidis, where, for Culture, Sports, Science and Technology Shinjuku-ku, Tokyo 162-8640, Japan; fax: +81- example, a 34-MDa R-plasmid of of Japan. 3-5285-1171; email: [email protected] incompatibility group N (IncN) (8) and a 36-MDa R-plasmid of IncX References (pDEP57) (6) were described. Both kinds of plasmids encoded ampicillin 1. National Institute of Infectious Diseases. resistance as well as that in this study, Salmonellosis in Japan as of June 2000. Infectious Agents Surveillance Report, but both were identified in PT6a iso- vol. 24;2003:162. Available from: URL: lates. Preliminary sequence data of http://idsc.nih.go.jp/iasr/24/282/tpc282 Factors Influencing the region of p981123 essential for .html Fluoroquinolone replication indicated a gene coding 2. Threlfall EJ. Epidemic Salmonella Typhimurium DT 104-a truly international Resistance for a protein similar to protein p1 of multiresistant clone. J Antimicrob R6K (IncX) plasmid (9), which sug- Chemother 2000;46:7–10. To the Editor: Recently, Scheld gests that p981123 may be related to 3. Matsune W, Ishikawa K, Hayashi KI, Tsuji summarized factors that he consid- pDEP57. However, the reactions to M, Izumiya H, Watanabe H. Molecular ered to have an influence on the effi- analysis of Salmonella Enteritidis isolates the typing phages in RDNC-a strains resistance to ampicillin and streptomycin cacy of fluoroquinolones (1). In the were different from those in PT6a. from three outbreaks of food poisoning in review, ciprofloxacin was presented Therefore, the R-plasmid in this study Shiga prefecture. Jpn J Infect Dis as the most active fluoroquinolone seems to have different features from 2001;54:111–3. against Pseudomonas aeruginosa

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 9, No. 12, December 2003 1651 LETTERS with MICs typically two- to eightfold rate than with ciprofloxacin. The review implied that, in gener- lower than those for levofloxacin, However, a recent study of P. aerugi- al, higher AUC24h/MIC ratios were moxifloxacin, or gatifloxacin. nosa isolated from cystic fibrosis associated with better patient out- However, because the National patients reported that fewer resistant comes. For S. pneumoniae, several Committee for Clinical Laboratory mutants were isolated after exposure pharmacodynamic studies have Standards (NCCLS) MIC interpreta- to levofloxacin (11 mutants) than to demonstrated that a target tive breakpoints are fluoroquinolone- ciprofloxacin (28 mutants) (8). AUC24h/MIC ratio of 30 to 35 for flu- specific, percent susceptibility is con- With regards to S. pneumoniae, the oroquinolones is the best correlate for sidered to be a better measure by review stated that in vitro studies have successful bacteriologic eradication, which to compare fluoroquinolone demonstrated that ciprofloxacin (1–4 clinical cure, and prevention of emer- activities. Our company has conduc- mg/L) and levofloxacin (1–2 mg/L) gence of resistance during therapy ted annual investigations called are not as active as moxifloxacin (5,9,17–19). Levofloxacin, gati- TRUST (Tracking Resistance in the (0.06–0.25 mg/L) and gatifloxacin floxacin, and moxifloxacin all achieve

United States Today) since 1996. (0.5–1 mg/L) against pneumococci. this AUC24h/MIC ratio (9). Zhanel et These surveillance studies have con- As with P. aeruginosa, fluoro- al. demonstrated that AUC24h/MIC sistently shown similar susceptibility quinolone comparisons against S. ratios above the target value of 30 to rates for levofloxacin (67.7% in 2002) pneumoniae should not be limited to 35 did not improve bacteriologic and ciprofloxacin (67.4% in 2002) MICs alone because pharmacokinetic eradication or reduce the emergence against P. aeruginosa (2,3). Both and pharmacodynamic characteristics of resistance (9). Moreover, no clini- agents show higher in vitro activity differ for each fluoroquinolone. cal data support the claim that higher against P. aeruginosa than gati- Pneumococcal time-kill studies with AUC24h/MIC ratios correlate with bet- floxacin and moxifloxacin (2–4). A levofloxacin, gatifloxacin, and moxi- ter patient outcomes. critique of antipseudomonal fluoro- floxacin in a pharmacodynamic The review discusses the question quinolone activity should also con- model have demonstrated that these of whether C-8-methoxyquinolones sider peak achievable fluoroquinolone three agents possess equal bacterici- (moxifloxacin and gatifloxacin) have levels at a site of infection, the area dal activity and are equally effective a lower propensity to select resistant under the serum concentration curve in preventing resistance development mutants of S. pneumoniae compared in 24 hours (AUC24h), and the because the lower in vitro MICs for with levofloxacin. Mutation preven- AUC24h/MIC ratio (5). At equivalent gatifloxacin and moxifloxacin were tion concentration is a theoretical lab- dosages for nosocomial pneumonia, offset by the higher serum and tissue oratory concept based on agar dilution levofloxacin (750 mg intravenously, levels of levofloxacin (9). In the same methodology, and no published data once daily) has a threefold higher study, ciprofloxacin did not exhibit have shown any clinical correlation peak serum level (Cmax) and threefold rapid killing and selected for resist- between this theory and clinical out- higher AUC24h than ciprofloxacin ance faster than the other three agents comes. NCCLS does not recommend (400 mg intravenously, every 8 hours) (9). TRUST and other U.S. surveil- agar dilution for susceptibility analy- (package inserts for Levaquin and lance studies, using the NCCLS-rec- sis of S. pneumoniae. Moreover, the Cipro). While certain P. aeruginosa ommended broth-dilution method, extremely low levels of resistance in isolates have lower ciprofloxacin than have shown that S. pneumoniae S. pneumoniae (<1%) after many levofloxacin MICs, the two fluoro- remain highly susceptible to lev- years of fluoroquinolone use do not quinolones have equivalent activity ofloxacin with resistance rates in the support the theory of mutation pre- against P. aeruginosa because of their United States of <1%; the MIC90 for vention concentration. The review did equivalent AUC24h /MIC ratios (6). levofloxacin in these studies has not reference an analysis of 16 peni- We agree strongly with Scheld’s sug- remained at 1 mg/L from 1997 cillin-resistant S. pneumoniae strains gestion that the fluoroquinolone used through 2002 (10–15). Further, lev- by Kolhepp et al. (20). In that broth- clinically should be the fluoro- ofloxacin, gatifloxacin, and moxi- dilution study, in vitro resistance quinolone tested by the laboratory and floxacin are equally effective in rates developed in a greater proportion of reported; surrogate testing of fluoro- of clinical cure and microbiologic strains exposed to gatifloxacin (11/16) quinolones may lead to major errors eradication of pneumococcal respira- and moxifloxacin (8/16) than to lev- in reporting, particularly for tory infections (16, and FDA website; ofloxacin (2/16). Similarly, in a study Enterobacteriaceae (2,3,7). available from: URL: http://www.fda. by Klepser et al. that used an in vitro The review also stated that lev- gov/cder/foi/nda/99/21061_Tequin.ht pharmacodynamic model, lev- ofloxacin-resistant strains of P. aerug- m and http://www.fda.gov/cder/foi/ ofloxacin was less likely than moxi- inosa emerge at a significantly higher nda/2001/21277_Avelox.htm) floxacin to select for resistant isolates

1652 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 9, No. 12, December 2003 LETTERS of S. pneumoniae; moreover, after 24 fluoroquinolones, such as ciproflo- rococci (31) and intestinal gram-nega- hours of exposure, levofloxacin MICs xacin and moxifloxacin, minimizes tive strains with increased fluoro- remained unchanged while moxi- fluoroquinolone resistance. Targeted quinolone resistance. floxacin MICs increased two- to fluoroquinolone therapy may in fact In conclusion, we believe that the eightfold (21). have adverse implications for the data we have briefly presented here Levofloxacin, gatifloxacin, and patient and for overall institutional supplements the previous discussion moxifloxacin all have susceptibility resistance patterns. For example, the by Scheld (1) and will help facilitate rates >99% for S. pneumoniae use of ciprofloxacin for urinary tract an improved understanding of the fac- (22,23). Although resistance is rare, infections exposes resident strepto- tors influencing the maintenance of considerable cross-resistance among cocci in the respiratory tract to an fluoroquinolone efficacy. fluoroquinolones is observed once agent that has demonstrated weaker two or more key mutations (e.g., Ser79 activity against pneumococci, thus Focus Technologies is the central in ParC, Ser81 in GyrA) are detected potentially selecting for pneumococ- testing laboratory for the TRUST antimi- (24,25). Using topoisomerase IV- cal resistance (9). Moreover, 20%- crobial susceptibility testing surveillance selecting fluoroquinolones (ciproflo- 35% of ciprofloxacin is excreted program, sponsored by Ortho-McNeil xacin and levofloxacin) in the same through the intestinal tract (Cipro Pharmaceutical. patient population as DNA gyrase- package insert), compared to 4% of selecting fluoroquinolones (gati- levofloxacin (Levaquin package floxacin and moxifloxacin) could insert). Studies have shown that Daniel F. Sahm,* potentially accelerate the develop- ciprofloxacin displays weaker in vitro Clyde Thornsberry,* ment of double mutants (ParC and activity (lower percentage of isolates Mark E. Jones,* GyrA) and clinically important class susceptible) than levofloxacin for sev- and James A. Karlowsky* resistance because selective pressure eral gram-negative enteric bacteria *Focus Technologies, Herndon, Virginia, would be applied to both enzyme tar- (2,3). Stepwise adaptive changes USA gets (26). towards fluoroquinolone resistance in The review stated that, since 1999, enteric bacteria may be selected by References at least 20 case reports of pulmonary fluoroquinolones with weaker in vitro 1. Scheld WM. Maintaining fluoroquinolone infection that did not respond to lev- activity and higher levels of exposure class efficacy: review of influencing fac- ofloxacin therapy have been pub- in the intestinal tract. Therefore, tors. Emerg Infect Dis 2003;9:1–9. 2. Karlowsky JA, Kelly LJ, Thornsberry C, lished. This number is remarkably ciprofloxacin would have a greater Jones ME, Evangelista AT, Critchley IA, et small considering that >250 million potential than levofloxacin for the al. Susceptibility to fluoroquinolones patients have been treated with lev- selection of resistant strains of intes- among commonly isolated Gram-negative ofloxacin worldwide. A number of the tinal gram-negative pathogens. A bacilli in 2000: TRUST and TSN data for the United States. Int J Antimicrob Agents treatment failures cited had documen- recent report stated that ciprofloxacin- 2002;19:21–31. tation of prior ciprofloxacin use and resistant Escherichia coli were isolat- 3. Blosser-Middleton RS, Sahm D, ciprofloxacin failure, and many iso- ed from the feces of 48% of patients Evangelista AT, Thornsberry C, Jones ME, lates were not tested for levofloxacin treated with ciprofloxacin for prostati- Critchley IA, Karlowsky JA. Antimicrobial susceptibilities of common pathogens caus- susceptibility before treatment (27). tis; before ciprofloxacin therapy, only ing nosocomial pneumonia: 2001–2002 We agree with the recommendation in ciprofloxacin-susceptible E. coli were TRUST surveillance. Annual Meeting the cited Davidson et al. reference: a isolated from the feces of these Infectious Disease Society of America, patient’s failure to respond to one flu- patients (28). Further, given that 25% 2002, abstract 71. 4. Milatovic D, Schmitz F-J, Brisse S, oroquinolone is sufficient reason not of moxifloxacin is excreted through Verhoef, Fluit AC. In vitro activities of to use other fluoroquinolones (27). the intestinal tract (Avelox package sitafloxacin (DU-6859a) and six other fluo- Isolated clinical failures will occur insert), the use of moxifloxacin for roquinolones against 8,796 clinical bacteri- with the use of any antimicrobial respiratory infections exposes bac- al isolates. Antimicrob Agents Chemother 2000;44:1102–7. agent when treating pneumococcal teria in the intestinal tract to a fluoro- 5. Craig WA. Does dose matter? Clin Infect pneumonia. quinolone with greater activity Dis 2001;33(Suppl 3):S233–7. The notion that fluoroquinolone against Bacteroides fragilis and other 6. MacGowan AP, Wootton M, Holt HA. The therapy can be “targeted” for an indi- intestinal anaerobes than levofloxacin antibacterial efficacy of levofloxacin and ciprofloxacin against Pseudomonas aerugi- cation requires challenge as fluoro- (29,30). Moxifloxacin has a greater nosa assessed by combining antibiotic quinolone therapy will always result potential than other fluoroquinolones exposure and bacterial susceptibility. J in systemic drug levels. Evidence to alter the normal intestinal flora and Antimicrob Chemother 1999;43:345–9. does not indicate that the use of two select for vancomycin-resistant ente-

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 9, No. 12, December 2003 1653 LETTERS

7. Sahm DF, Thornsberry C, Jones ME, of the fluoroquinolones: focus on respirato- 26. Davies TA, Evangelista A, Pfleger S, Bush Blosser R, Critchley IA, Evangelista AT, ry infections. Drugs 2002;62:13–59. K, Sahm DF, Goldschmidt R. Prevalence of Karlowsky JA. Antimicrobial susceptibility 17. Lacey MK, Lu W, Xu X, Tessier PR, single mutations in topoisomerase type II of Enterobacteriaceae and Pseudomonas Nicolau DP, Quintiliani R, Nightingale CH. genes among levofloxacin-susceptible clin- aeruginosa from inpatient infections in the Pharmacodynamic comparisons of lev- ical isolates of Streptococcus pneumoniae U.S.: 1999–2002 TRUST surveillance. ofloxacin, ciprofloxacin, and ampicillin isolated in the United States in 1992–1996 Critical Care Congress, 2003, Abstract against Streptococcus pneumoniae in an in and 1999–2000. Antimicrob Agents 22015. vitro model of infection. Antimicrob Agents Chemother 2002;46:119–24. 8. Gillespie T, Masterton RG. Investigation Chemother 1999;43:672–7. 27. Davidson R, Covalcanti R, Brunton JL, into the selection frequency of resistant 18. Nightingale CH, Grant EM, Quintiliani R. Bast DI, de Azavedo JC, Kibsey P, et al. mutants and the bacterial kill rate by lev- Pharmacodynamics and pharmacokinetics Resistance to levofloxacin and failure of ofloxacin and ciprofloxacin in non-mucoid of levofloxacin. Chemotherapy 2000;46 treatment of pneumococcal pneumonia. N Pseudomonas aeruginosa isolates from (Suppl 1):6–14. Engl J Med 2002;346:747–50. cystic fibrosis patients. Int J Antimicrob 19. Ambrose PG, Grasela DM, Grasela TH, 28. Horcajada JP, Vila J, Moreno-Martínez A, Agents 2002;19:377–82. Passarell J, Mayer HB, Pierce PF. Ruiz J, Martínez J, Sánchez M, Soriano E, 9. Zhanel GG, Walters M, Laing N, Hoban DJ. Pharmacodynamics of fluoroquinolones et al. Molecular epidemiology and evolu- In vitro pharmacodynamic modeling simu- against Streptococcus pneumoniae in tion of resistance to quinolones in lating free serum concentrations of fluoro- patients with community-acquired respira- Escherichia coli after prolonged adminis- quinolones against multidrug-resistant tory tract infections. Antimicrob Agents tration of ciprofloxacin in patients with pro- Streptococcus pneumoniae. J Antimicrob Chemother 2001;45:2793–7. statitis. J Antimicrob Chemother Chemother 2001;47:435–40. 20. Kolhepp SJ, Grunkemeier G, Leggett JE, 2002;49:55–9. 10. Thornsberry C, Ogilvie PT, Holley HP Jr, Dworkin RJ, Slaughter SE, Gilbert DN. 29. Hoellman DB, Kelly LM, Jacobs MR, Sahm DF. Survey of susceptibilities of Phenotypic resistance of penicillin-suscep- Appelbaum PC. Comparative antianaerobic Streptococcus pneumoniae, Haemophilus tible and penicillin-resistant Streptococcus activity of BMS 284756. Antimicrob influenzae, and Moraxella catarrhalis iso- pneumoniae after single and multiple in Agents Chemother 2001;45:589–92. lates to 26 antimicrobial agents: a prospec- vitro exposures to ciprofloxacin, gati- 30. Ednie LM, Jacobs, Appelbaum PC. tive U.S. study. Antimicrob Agents floxacin, levofloxacin, moxifloxacin, and Activities of gatifloxacin compared to those Chemother 1999;43:2612–23. trovofloxacin. Annual Meeting Infectious of seven other agents against anaerobic 11. Biedenbach DJ, Barrett MS, Croco MA, Diseases Society of America, 2000, organisms. Antimicrob Agents Chemother Jones RN. Bay 12-8039, a novel fluoro- Abstract 97. 1998;42: 2459–62. quinolone, activity against important respi- 21. Klepser M, Ernst E, Petzold CR, Rhomberg 31. Zhanel GG, Laing NM, DeCorby M, Nichol ratory tract pathogens. Diagn Microbiol P, Doern GV. Comparative bactericidal KA, Hoban DJ. Pharmacodynamic activity Infect Dis 1998;31:45–50. activities of ciprofloxacin, clinafloxacin, of fluoroquinolones in a mixed infection 12. Jones RN, Pfaller MA. In vitro activity of grepafloxacin, levofloxacin, moxifloxacin, simulationg an artificial bowel: effect of newer fluoroquinolones for respiratory tract and trovafloxacin against Streptococcus eradicating Bacteroides fragilis. American infections and emerging patterns of antimi- pneumoniae in a dynamic in vitro model. Society for Microbiology, 2002, Abstract crobial resistance data from the Sentry Antimicrob Agents Chemother 2001;45: A-145. antimicrobial surveillance program. Clin 673–8. Infect Dis 2000;31(Suppl 2):S16–23. 22. Low D, de Azavedo J, Weiss K, Mazzulli T, Address for correspondence: James A. 13. Doern GV, Heilmann KP, Huynh HK, Kuhn M, Church D, et al. Antimicrobial Karlowsky, Focus Technologies, 13665 Dulles Rhomberg PR, Coffman SL, Brueggemann resistance among clinical isolates of AB. Antimicrobial resistance among clini- Streptococcus pneumoniae in Canada dur- Technology Drive, Suite 200, Herndon, VA cal isolates of Streptococcus pneumoniae in ing 2000. Antimicrob Agents Chemother 20171-4603, USA; fax: (703) 480-2654; email: the United States during 1999-2000, includ- 2002;46:1295–301. [email protected] ing a comparison of resistance rates since 23. Brueggemann AB, Coffman SL, Rhomberg 1994–1995. Antimicrob Agents Chemother P, Huynh H, Almer L, Nilius A, et al. 2001;45:1721–9. Fluoroquinolone resistance in Strepto- 14. Thornsberry C, Sahm DF, Kelly LJ, coccus pneumoniae in United States since Critchley IA, Jones ME, Evangelista AT, et 1994–1995. Antimicrob Agents Chemother al. Regional trends in antimicrobial resist- 2002;46:680–8. ance among clinical isolates of 24. Evangelista AT, Loeloff M, Pfelger S, Streptococcus pneumoniae, Haemophilus Davies T, Bush K, Mauriz Y, et al. Cross- International influenzae, and Moraxella catarrhalis in resistance among fluoroquinolone-resistant the United States: results from the TRUST clinical isolates of Streptococcus pneumo- Travel and Sexually surveillance program, 1999-2000. Clin niae. J Antimicrob Chemother 2001;47 Infect Dis 2002;34(Suppl 1):S4–16. (Suppl 1):29, Abstract P50. Transmitted 15. Sahm DF, Thornsberry C, Jones ME, 25. Davies TA, Pfleger S, Goldschmidt R, Bush Disease Blosser RS, Critchley IA, Evangelista AT, K, Sahm DF, Evangelista AT. et al. Correlation of antimicrobial resistance Characterization of U.S. clinical Strepto- To the Editor: Recent articles in among Streptococcus pneumoniae in the coccus pneumoniae strains from U.S.: 2001–2002 TRUST surveillance. 2000–2001 that are cross-resistant to the professional literature (1–3) have Interscience Conference on Antimicrobial ciprofloxacin, gatifloxacin, levofloxacin, offered advice regarding the impor- Agents and Chemotherapy, 2002, Abstract and moxifloxacin. Annual Meeting tance of taking a careful travel history, C2-1640. Infectious Disease Society of America particularly in this time of unprece- 16. Zhanel GG, Ennis K, Vercaigne L, Walkty 2002, Abstract 78. A, Gin AS, Embil J, et al. A critical review dented levels of international travel

1654 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 9, No. 12, December 2003