以前、グラム陰性桿菌の菌血症や耐性について、ブログに記載しました。グラム陰性桿菌の菌血症 - akinohanayukiの日記 〜 common diseaseと薬学を中心に〜
In some cases, however, a more broadly active antibiotic may be the drug of choice for directed therapy even if the organism tests susceptible to an agent with a narrower spectrum. As an example, whether organisms that produce an inducible chromosomal AmpC beta-lactamase (eg, Enterobacter, Serratia, Citrobacter, indole-positive Proteus, Providencia, Morganella) can be successfully treated with a third-generation cephalosporin or a beta-lactam/beta-lactamase inhibitor remains uncertain. Most of this concern is theoretical, supported largely by in vitro data with limited information on whether this phenomenon actually leads to clinical failure in patients. Furthermore, despite the presence of AmpC beta-lactamases in several human pathogens, relevant clinical data are limited to observational studies in which a higher risk of clinical failure has been demonstrated with bacteremia or meningitis due to Enterobacter spp. treated with third-generation cephalosporins and in a small number of patients in whom Serratia and Citrobacter developed resistance during therapy. For most patients with gram-negative bacteremia due to AmpC producers, we believe that following susceptibility data and using a beta-lactam drug to which the organisms test susceptible is adequate. However, we prefer treatment with cefepime or a carbapenem when the primary infection is in the central nervous system (CNS) or another sequestered site which may not receive adequate drug penetration or the planned treatment course is long (>14 days). Repeat susceptibility testing for subsequently identified isolates is necessary to detect evidence of emerging resistance during the course of therapy． For non-CNS infections with AmpC producers that have confirmed susceptibility in vitro, use of quinolones avoids beta-lactamase induction and remains a good option for directed treatment of bloodstream infections due to excellent bioavailability.
Although virtually all Gram negative bacilli possess a chromosomal beta-lactamase gene, certain species express insignificant amounts of this enzyme and their susceptibility to beta-lactams is largely determined by plasmid-mediated beta-lactamases and antibiotic permeability. These include E. coli, Proteus mirabilis, Salmonella, Shigella, and H. influenzae. Klebsiella pneumoniae produces a chromosomal beta-lactamase that is primarily a penicillinase; thus, these strains are frequently more susceptible to the cephalosporins. The last group of species within the Enterobacteriaceae, including Enterobacter, indole-positive Proteus, Serratia, and Citrobacter, produce an inducible chromosomal beta-lactamase that may be difficult to detect on initial susceptibility testing but that can mediate resistance to all currently available beta-lactams with the exception of the carbapenems and perhaps cefepime. In addition to inducible production of this chromosomal enzyme, these species may give rise to regulatory mutants that are "derepressed" and produce high levels of this broad-spectrum chromosomal enzyme constitutively.
Serratia species are intrinsically resistant to ampicillin, amoxicillin, ampicillin-sulbactam, amoxicillin-clavulanate, narrow-spectrum cephalosporins, cephamycins, cefuroxime, macrolides, tetracyclines, and nitrofurantoin, and colistin. Additionally, Serratia species have the potential to harbor multidrug resistance mechanisms (such as AmpC or extended-spectrum beta-lactamases and carbapenemases) and to develop resistance to broad-spectrum beta-lactams during therapy.
The risk of emergence of AmpC-mediated resistance during therapy is likely high with central nervous system infections, infections in sequestered sites that require prolonged antibiotic therapy, and cases in which infected material cannot be removed or adequately débrided. For patients with such infections, we suggest not using a third-generation cephalosporin (eg, ceftriaxone or ceftazidime) even if the isolate tests susceptible (Grade 2C). Instead, for central nervous system infections, we favor the use of a carbapenem (meropenem) or cefepime. For other infections in sequestered sites, fluoroquinolones and trimethoprim-sulfamethoxazole are likely appropriate.
High level expression of AmpC beta-lactamase or an extended-spectrum beta-lactamase further limits treatment options, generally to the carbapenem class. However, cefepime and agents outside the beta-lactam class may still be effective in isolates that produce AmpC if testing demonstrated susceptibility to those agents.
Duration of therapy depends on the site of infection and the patient’s clinical response. Culture and susceptibility testing should be repeated and therapy adjusted accordingly for patients who do not clinically improve on cephalosporin therapy because of the potential to select for AmpC beta-lactamase producing isolates.