correspondence - Wiley Online Library

0 downloads 0 Views 48KB Size Report
case for carboxy-ureidopenicillins, third-gen- eration cephalosporins and aztreonam. These antibiotics are active against inducible bacteria in the absence of ...
CORRESPONDENCE

Cephalosporinase induction and cephalosporin resistance: a longstanding misinterpretation

For more than 20 years, and until very recently, it has been suggested that ‘all Enterobacter spp., Citrobacter freundii, Serratia spp., Providencia spp. and Morganella morganii should be reported resistant to all third-generation cephalosporins’, with the exception of urinary tract infection isolates [1]. The alleged reason is that these ampC-inducible species yield ‘derepressed’ mutants, producing high levels of AmpC b-lactamase which are responsible for clinical failure. Other authors have recommended screening for the inducible AmpC b-lactamase and reporting as resistant only those strains giving a positive result. The aim of this letter is to analyze the genetic background, the in vitro results, and the clinical data supporting such an approach. Confusion between induction and selection of resistant mutants is aggravated by the very common misuse of the word ‘induction’ in the literature. Induction of the AmpC chromosomal b-lactamase: virtually all the above-mentioned bacteria (but also Pseudomonas aeruginosa and other nonfermentative Gram-negative bacteria) possess an inducible b-lactamase with high affinity for cephalosporins, produced at low levels [2–4]. It is a characteristic trait of these bacterial species; therefore, reports that more than 20% of Enterobacter and Morganella, and more than 60% of Serratia strains are, not inducible can only be due to inadequate testing methods or misidentification [5]. In the presence of a b-lactam which interacts with the penicillin-binding proteins (PBPs) of the bacteria, the resulting increase in peptidoglycan precursors will trigger the activity of AmpG, a permease which acts as a transmembrane signal transducer. AmpG will overcome the joint control of ampD and ampE and promote the conversion of the repressed ampR into an activated form (inactivation or loss of ampE or ampG results in a total absence of induction). The activated ampR will stimulate the expression of ampC, producing large amounts of the AmpC chromosomal b-lactamase. However, as soon as the inducer is removed, ampC expression will return to normal. b-Lactams can be divided into four categories on the basis of their induction potential and their

stability in the presence of increased amounts of b-lactamase produced. (1) Inducer and unstable: aminopenicillins, firstgeneration cephalosporins, cephoxitin, cephotetan, and also clavulanic acid. These antibiotics are rapidly inactivated by the increased b-lactamase induced. (2) Inducer but stable: one example is imipenem, which remains active as long as another mechanism of resistance (decreased permeability) is not associated. (3) Non-inducer and limited stability: this is the case for carboxy-ureidopenicillins, third-generation cephalosporins and aztreonam. These antibiotics are active against inducible bacteria in the absence of another inducer b-lactam. (4) Non-inducer and relatively stable, such as cefpirome and cefepime, these antibiotics do not have the stability of imipenem and their activity is affected in case of stable overproduction resulting from mutation. The major consequence of induction is the wildtype pattern of the involved species towards b-lactam antibiotics, i.e. resistance to aminopenicillins (alone or combined with clavulanate or sulbactam), first-generation cephalosporins and sensitivity to carboxy- and ureidopenicillins, thirdgeneration cephalosporins, aztreonam, cefepime, cefpirome and imipenem. Induction is an in vitro phenomenon with no clinical impact; the only risk would be the very unlikely combination of a strong inducer such as cefoxitin with a third-generation cephalosporin. Selection of resistant mutants (‘derepression’): in contrast to induction, which is a transient phenomenon, selection of resistant mutants producing very high levels of b-lactamase has been observed in vitro and in vivo [1,4,6–10]. These mutants, observed with a frequency of 106108 of wild-type bacterial cells, are resistant to third-generation cephalosporins and aztreonam, with minimum inhibitory concentrations (MICs) of 32–256 mg/L and usually also to carboxy- and ureidopenicillins. All active b-lactams, except imipenem, can select such resistant mutants, but the less stable, poor inducer compounds are the best selectors. Indeed, for these compounds there is a high increase in MICs observed between the wild-type and the derepressed mutants which often spans the antibiotic concentration, resulting in killing the former and selecting the latter.

ß 2002 Copyright by the European Society of Clinical Microbiology and Infectious Diseases

824 Clinical Microbiology and Infection, Volume 8 Number 12, December 2002

The selection of resistant mutants is antibiotic-, inoculum- and concentration-dependent and also species- and strain-dependent. The frequency of selection of resistant mutants is lower in Serratia than in Enterobacter; in addition, the MICs attained are much lower, especially with ceftazidime: in one study, the MICs were 256 and 64 mg/mL for cefotaxime and ceftazidime, respectively, in E. cloacae, versus 8 and 1 mg/mL, respectively, in Serratia [4]. This example indicates that not all bacteria and antibiotics have the same level of risk. From a genetic point of view, induction and the selection of resistant mutants follow different pathways. The resistant mutant pre-exists in the susceptible bacterial population; it is neither ‘induced’ nor obtained by a mutagenic effect of the antibiotic. It essentially involves a modification of the ampD gene, which can no longer repress ampR, resulting in stable stimulation of ampC. Hence, the ability of an antibiotic to select resistant mutants is unrelated to its ability to induce the AmpC b-lactamase [2,3,7]. The so-called third-generation cephalosporins have been successfully used during the last 25 years to treat a wide array of infections. These antibiotics were considered as a major breakthrough, extending the activity of cephalosporins to all enterobacterial species and Pseudomonas, together with very favorable pharmacokinetic parameters and very low toxicity. A number of failures have been reported, but with a highly variable incidence [7–10]; the highest incidence of emergence of resistance during therapy (up to 20%), reported in reference [1], was observed in severely ill patients with Enterobacter bacteremia, but the figures were small (six out of 31 patients) [8]. Moreover, at least two of these strains, with MICs of 8 mg/mL, were clearly not (fully) susceptible and the patients should have been treated with other antibiotics. Resistance to cephalosporins is selected stepwise and involves various mechanisms (not only ‘derepression’). Low-level resistant mutants (often due to decreased permeability) are more likely to mutate to a higher level of resistance; it is the role of the clinical microbiologist to recognize such strains and inform the clinician that, despite the classification of the strain as ‘susceptible’ (by the NCCLS but not by the BSAC or the French Antibiogram Committee), there is an increased risk of clinical failure. In several other studies, the rate of clinical failure was lower than the incidence of resis-

tant mutant selection (6% versus 28% in one study [7] and 5.2% versus 9.9% in another study [10]). In a study involving 392 patients, selection of resistant mutants occurred in 6.1% of patients infected by ‘inducible’ bacteria and treated with third-generation cephalosporins or aztreonam. The failure rate was 8.4% for Enterobacter spp., 5% for Citrobacter, 4.8% for Serratia and 0% for Morganella and Providencia. More unexpected in this study was the emergence of resistance in bacteremic patients: none of the 305 patients treated [9]. A higher failure rate is reported in a very recent paper, but the MICs against the susceptible and resistant isolates are not reported [11]. Administration of another antibiotic (an aminoglycoside or fluoroquinolone) may decrease, but not totally eliminate, the emergence of resistant strains, probably owing to differences in their pharmacokinetics [8]. Emergence of resistance to third-generation cephalosporins in Enterobacter, C. freundii and Serratia has to be compared to the emergence of resistance to other antibiotics such as aminoglycosides and fluoroquinolones. In one review, resistant strains emerged in 9.9% of patients treated with cephalosporins, 5.9% of patients treated with ciprofloxacin and 13.8% of patients treated with aminoglycosides for infections due to the abovementioned bacteria [10]. In the same study, P. aeruginosa resistance developed in 24.5% of patients treated with ciprofloxacin and 21.1% of patients treated with aminoglycosides. In other studies, P. aeruginosa resistance emerged in 16.9% of patients treated with aminoglycosides [12] and 29.7% [12] to 34.7% [9] of patients treated with imipenem. Selection of resistant mutants during treatment is a serious event, but is offset by the overall cure rates and by the advantages of using third-generation cephalosporins, and must also be considered in the light of the therapeutic alternatives. The ideal compound does not exist cefepime and, to a lesser extent cefpirome, have a much lower propensity to select resistant mutants, but it can happen [6,13]. In an excellent in vitro study, resistance to cefepime and cefpirome was easily selected in E. cloacae after only one or two steps, with MICs up to 32 and 128 mg/mL, respectively [6]. Moreover, these antibiotics readily select mutants with altered permeability (80% of clones versus 10% selected with ceftriaxone or ceftazi-

ß 2002 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 8, 823–825

Correspondence 825

dime) [6]. These clones are resistant not only to the selecting agent but also to unrelated antibiotics such as tetracyclines, chloramphenicol, trimethoprim and especially fluoroquinolones, curtailing the use of such compounds. These antibiotics may also select resistant Klebsiella oxytoca (mutation of the chromosomal b-lactamase) or Escherichia coli harboring plasmidmediated b-lactamases such as Oxa-1, which affect cefepime more than ceftazidime. Cefepime and cefpirome are recommended for nosocomial bacteremia and pneumonia (particularly in intensive care units where Enterobacter spp. represent a significant risk), but should not replace third-generation cephalosporins for other infections. Imipenem is clearly less active, more toxic and more difficult to administer; selection of resistance may occur in the same species through different mechanisms, curtailing the future use of other blactams. A final factor which has to be taken into account is the cost of other antibiotics in comparison to third-generation cephalosporins. The suggestion that all enterobacterial species harbouring an inducible AmpC b-lactamase should be considered as resistant to third-generation cephalosporins is neither accurate nor realistic. Alarming data from a single study of Enterobacter spp. are not very helpful; the clinician needs more objective information on the advantages and risks of using third-generation cephalosporins or alternatives. A last point: if enterobacteria with inducible cephalosporinase were considered as resistant to third-generation cephalosporins, we should also consider, based on similar observations of potential selection of resistant mutants, that all P. aeruginosa are resistant to ceftazidime and all Mycobacterium tuberculosis are resistant to streptomycin, isoniazid, rifamycin, or fluoroquinolones – which is a nonsense. F. W. Goldstein Hospital Saint Joseph, 185 rue Raymond Losserand, 75014, Paris, France Tel: þ33 1 44 12 36 50 Fax: þ33 1 44 12 36 85

REFERENCES 1. Livermore DM, Brown Derek FJ. Detection of b-lactamase-mediated resistance. J Antimicrob Chemother 2001; 48 (Suppl. 1): 59–4. 2. Bennett PM, Chopra I. Molecular basis of blactamase induction in bacteria. Antimicrob Agents Chemother 1993; 37: 153–8. 3. Honore´ N, Nicolas MH, Cole ST. Regulation of enterobacterial cephalosporinase production: the role of a membrane-bound sensory transducer. Mol Microbiol 1989; 3: 1121–30. 4. Livermore DM. b-Lactamases in laboratory and clinical resistance. Clin Microbiol Rev 1995; 8: 557–84. 5. European study group on antibiotic resistance. Incidence of inducible beta-lactamases in Gramnegative septicemia isolates from twenty-nine European laboratories. Eur J Clin Microbiol 1987; 6: 460–6. 6. Fung Tomc JC, Gradelski E, Huczko E, Dougherty TJ, Kessler RE, Bonner DP. Differences in the resistant variants of Enterobacter cloacae selected by extended-spectrum cephalosporins. Antimicrob Agents Chemother 1996; 40: 1289–93. 7. Sanders CC, Sanders WE Jr Microbial resistance to newer generation b-lactam antimicrobials: Clinical and laboratory implications. J Infect Dis 1985; 151(Suppl. 1): 399–6. 8. Chow JW, Fine MJ, Shlaes DM et al. Enterobacter bacteremia: clinical features and emergence of antibiotic resistance during therapy. Ann Intern Med 1991; 115: 585–90. 9. Fish DN, Piscitelli SC, Danziger LH. Development of resistance during antimicrobial therapy: a review of antibiotic classes and patient characteristics in 173 studies. Pharmacotherapeutics 1995; 15: 279–91. 10. Milatovic D, Braveny I. Development of resistance during antibiotic therapy. Eur J Clin Microbiol 1987; 6: 234–44. 11. Kaye KS, Cosgrove S, Harris A, Eliopoulos GM, Carmeli Y. Risk factors for emergence of resistance to broad-spectrum cephalosporins among Enterobacter spp. Antimicrob Agents Chemother 2001; 45: 2628–30. 12. Carmeli Y, Troillet N, Eliopoulos GM, Samore MH. Emergence of antibiotic-resistant Pseudomonas aeruginosa: Comparison of risks associated with different antipseudomonal agents. Antimicrob Agents Chemother 1999; 43: 1379–82. 13. Sanders WE, Jr, Tenney JH, Kessler RE. Efficacy of cefepime in the treatment of infections due to multiply resistant Enterobacter species. Clin Infect Dis 1996; 23: 454–61.

ß 2002 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 8, 823–825