Optimizing Antimicrobial Therapy for Serious Infections in the Critically Ill

Critically ill patients pose many challenges when they develop serious infections. Antimicrobial choices are typically limited by multidrug resistance, allergies, or compromised renal/hepatic function. Dosing regimens, typically devised from studies in healthy volunteers, may be inadequate for individuals with major fluid shifts or other issues altering their pharmacokinetics. Clearly, the stakes in treating infections are also higher in critically ill patients with limited physiological reserve. This issue describes research that gives clinicians insight into the optimal antimicrobial regimens for therapy of critically ill patients with serious infections. The first four articles deal with optimizing the dosing of β-lactam antibiotics, fluoroquinolones, aminoglycosides, and polymyxins. The remaining eight articles deal with problem pathogens.

Jason Roberts and Jeffrey Lipman open this issue with their review on optimization of therapy with β-lactam antibiotics for critically ill patients. As is well known, the key determinant in efficacy of this antibiotic class is maintaining the concentration of the antibiotic above the minimal inhibitory concentration (MIC) for a prolonged duration. They point out that this is best achieved by use of continuous infusions of the antibiotics. Some antibiotics (e.g., carbapenems) do not have stability at room temperature for 24 hours, so these may be optimally administered by extended infusions (i.e., administration of each dose over 3 to 4 hours). This may also be the preferred approach in certain patients receiving infusions of numerous other medications.

Jenny Yang, Brian Tsuji, and Alan Forrest emphasize the importance of individualizing therapy with fluoroquinolones in critically ill patients. If this is done, it is possible for emergence of resistance to be minimized and bacterial killing to be maximized. This may, for example, mean use of ciprofloxacin 400 mg every 8 hours intravenously rather than the standard 400 mg every 12 hours.

Rhonda Rea and Blair Capitano discuss the optimization of aminoglycoside dosing. Despite the known toxicity of aminoglycosides, members of this class are still useful, especially in empirical regimens where multiresistant gram-negative pathogens such as Pseudomonas aeruginosa may be encountered. Unlike the β-lactam antibiotics, the aminoglycosides display concentration-dependent killing. The target for use of the aminoglycosides is to achieve a ratio of peak concentration to MIC of at least 10 with the first dose of the antibiotic. This may be best achieved by giving an initial dose of 7 mg/kg of gentamicin or tobramycin or 15 mg/kg of amikacin.

Unfortunately, many physicians caring for critically ill patients have needed to resort to use of colistin or polymyxin B for therapy of multidrug-resistant Acinetobacter, Pseudomonas aeruginosa, or Klebsiella pneumoniae. Roger Nation and Jian Li illustrate the difficulties posed in use of colistin or polymyxin B in critically ill patients. Foremost among these is the absence of guidance in the product information for these antibiotics with respect to their use in patients receiving continuous renal replacement therapy or hemodialysis. Both colistin and polymyxin B were developed long before modern concepts in pharmacokinetics/pharmacodynamics of antibiotics were elucidated. Drs. Nation and Li illustrate which pieces of data are required in order for rational dosing regimens of these drugs to be established.

Staphylococcus aureus infections, especially those due to methicillin-resistant Staphylococcus aureus (MRSA), are of huge importance in the intensive care unit (ICU). MRSA may arise in the community, in nursing homes, or in hospital wards, prompting admission to the ICU. Conversely, MRSA infections may arise in the ICU in critically ill patients admitted for other reasons. MRSA pneumonia is of particularly high mortality. Lee Skrupky, Scott Micek, and Marin Kollef describe the therapies available to treat MRSA pneumonia.

In the following article, Sara Cosgrove and Vance Fowler describe new innovations in the therapy of bloodstream infection with S. aureus, with a special emphasis on MRSA. They point out the prime importance of removal of infected foreign bodies and debridement of infectious foci. They also discuss the importance of exclusion of occult infective endocarditis. They review the role of newer agents such as daptomycin in therapy of this infection.

Peter Linden outlines the problems posed by infections due to vancomycin resistant Enterococcus faecium (VRE). Foremost is the decision as to whether the organism is merely colonizing the patient or is causing true infection. If a true infection has been identified, it is important to remove foreign bodies and attempt percutaneous or surgical drainage of closed space infections. Failure to do so may increase the risk both of failure of treatment and of development of resistance during therapy. This has been well described with use of linezolid.

With regard to therapy of multiply resistant gram-negative infections, resistant Enterobacteriaceae plus resistant Pseudomonas aeruginosa, Acinetobacter baumannii, and Stenotrophomonas maltophilia are covered in this issue. Andrea Endimiani and I describe treatment options for extended-spectrum β-lactamase (ESBL) producing Klebsiella pneumoniae. Carbapenems are the treatment of choice for serious infections with ESBL-producing organisms. However, many institutions have observed the advent of KPC-producing organisms, which are resistant to carbapenems.

Jon Iredell has presented an interesting overview of treatment of serious P. aeruginosa infections. He has tackled the thorny issue of whether combination therapy should be routinely used for this infection. Given the frequency with which resistance to the core antipseudomonal β-lactam antibiotics occurs, he concludes that it is quite reasonable to use combination therapy in the empirical setting.

Multidrug-resistant Acinetobacter baumannii has become an entrenched, problematic pathogen in many ICUs. Anton Peleg reviews treatment options. Meropenem- or sulbactam-based regimens may be appropriate for patients infected with strains susceptible to these antibiotics. However, many clinicians now face carbapenem- and sulbactam-resistant isolates. Amikacin, polymyxins (colistin or polymyxin B), or tigecycline may be the only treatment options with in vitro activity against the infecting organism. Clinical experience with tigecycline is limited at the present time, and this antibiotic is not approved by the U.S. Food and Drug Administration (FDA) for treatment of Acinetobacter infections. Furthermore, tigecycline achieves insufficient blood concentrations for it to be a reliable option in the treatment of bloodstream infection with A. baumannii. Development of resistance during therapy is also an emerging concern.

Robert Muder follows up with a clear description of the problems posed by another gram-negative bacillus, Stenotrophomonas maltophilia. This organism is resistant to multiple antibiotics, including most β-lactam antibiotics, including carbapenems, and aminoglycosides. Thus it is typically not covered by commonly used empirical antibiotic regimens. Trimethoprim-sulfamethoxazole is the antimicrobial agent of choice for this pathogen but may be best used as part of a synergistic combination regimen for patients with life-threatening infections due to S. maltophilia.

In the final article of this issue, E. Geoffrey Playford and Tania Sorrell outline advances in the management of invasive candidal infections in critically ill patients. Physicians now have three options available to treat serious Candida infections: azoles, echinocandins, and amphotericin preparations. This is important given that resistance is developing to standard agents such as fluconazole. Despite the availability of these antifungal options, physicians are still limited by the lack of early clinical or microbiological markers of invasive candidal infections.

Although this issue deals with therapy of serious infections with multidrug-resistant organisms, it is important for us to briefly consider prevention and control of these organisms. Classical infection-control precautions remain of paramount importance. At its most basic level, physicians in critical care medicine should insist on hand hygiene for every person entering and leaving the room of patients in ICUs. With many of the organisms described above, including the resistant gram-negative bacilli, contact isolation precautions (implying use of gloves and gowns while in the patient's room) are necessary. Although the decision to start antibiotic use in a critically ill patient is a difficult one to control, it is much easier to curb antibiotic duration. Several studies have now addressed duration of antibiotic use for conditions such as ventilator-associated pneumonia. Similar studies for other conditions should be encouraged.

I trust that this issue provides food for thought for individuals involved in the care of critically ill patients. Indeed, I believe there are several items discussed herein that may lead to changes in our practice. It may also be a stimulus for further outcomes-based research in this area.