28
Antibiotic Therapy of Multidrug-Resistant
Pseudomonas aeruginosa, Klebsiella
pneumoniae, and Acinetobacter
baumannii in Critical Care
Burke A. Cunha
Infectious Disease Division, Winthrop-University Hospital, Mineola, New York,
and State University of New York School of Medicine, Stony Brook, New York, U.S.A.INTRODUCTION
Multidrug-ResistantP. aeruginosa, K. pneumoniae,andA. baumannii
Therapeutically, the most problematic microorganisms encountered in daily practice in critical
care areP. aeruginosa, K. pneumoniae, andA. baumannii. The aerobic gram-negative bacilli
(GNBs) are usually sensitive to a variety of antibiotics, but some strains may become resistant
to multiple antibiotics from different classes and are then considered to be multidrug resistant
(MDR) isolates. Antibiotic stance among these three species may be related to a mutation that
causes resistances or may be induced by certain antibiotics-mediated resistance, or may be
clonally spread throughout the critical care unit (CCU) or the ward hospital or even the region.
The clonal dissemination of MDR GNBs in the CCU and beyond is not caused or related to
antibiotic use. Clonally derived MRD GNB isolates may be spread extensively if not limited by
effective infection-control containment measures. Clonal spread of MDR GNBs may result in a
widespread resistance within an institution and not related to antibiotic usage patterns.
Although problematic for individual patients colonized/infected with MDR/GNBs, contain-
ment of the clonally derived isolate to a single patient limits the magnitude of potential
resistance problems in the CCU and institution.
The other type of resistance which is not caused by mutation and spread by
dissemination of MDR clonal isolates is that associated with antibiotic use. It is a common
clinical misconception that antibiotics have the same resistance potential or that the resistance
potential is related to antibiotic class. Antibiotic resistance is not related to volume of use, i.e.,
“antibiotic tonnage,” antibiotic class, or duration of time that the drug has been on the market
or the duration of postmarket exposure, i.e., years available for general use. Attempts have
been made to correlate structure–activity relationships with antibiotic resistance with different
classes of antibiotics. This approach applies to relatively few antibiotic aminoglycosides,
but not to the majority of antibiotics in other antibiotic classes. A historical approach to
understanding antibiotic-associated resistance from a clinical standpoint indicates that some
antibiotics are more likely to cause resistance than others. These antibiotics may be termed
“high-resistance potential” antibiotics indicating the resistance potential is not necessarily high
in terms of percentage but relatively higher than those with a “low-resistance potential.”
Antibiotics referred to as low-resistance potential antibiotics are those which when used in
high volume over extended periods of time have not been associated with acquired resistance
to various microorganisms. While antibiotics should not be used thoughtlessly, all other things
being equal, it is always preferable to use an antibiotic with a low resistance potential, in
preference to one with a high resistance potential. Common examples of low-resistance
potential antibiotics are amikacin among the aminoglycosides; meropenem, ertapenem, and
doripenem among the carbapenems; doxycycline and monocycline among the tetracyclines;
cefazolin, cephalexin, and cefprozil among the first-generation cephalosporins; cefoxitin and
cefotetan among the second-generation cephalosporins; cefotaxime, ceftizoxime, cefoperazone,
and ceftriaxone among the third-generation cephalosporins; cefepime among the fourth-
generation cephalosporins, aztreonam among the monobactams; piperacillin among the
anti-pseudomonal penicillins; levofloxacin and moxifloxacin among the quinolones;