De-escalation antimicrobial chemotherapy in critically III patients: pros and cons.

In spite of advances in critical care, nosocomial infections still have a considerable impact on Intensive Care Unit (ICU) and hospital length of stay, mortality and costs. Several authors suggest that antibiotic therapy should be instituted as soon as sepsis is suspected in critically patients. Over the last two decades the rates of occurrence for pathogens have significantly changed under selective pressure from broad-spectrum antimicrobial therapy. Shifts from predominance of gram-negative to gram-positive organisms and outbreaks of resistant pathogens address the need for appropriate empirical regimens. Agents such as ceftazidime, imipenem and, more recently, meropenem and tazobactam have been used successfully as monotherapy. Two different clinical trials have reported that meropenem monotherapy is significantly more effective than ceftazidime-based therapy. Because of the outbreak of methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis, some investigators suggest adding a glycopeptide to beta-lactamase inhibitor and carbapenem as initial empirical therapy. Such a regimen should be administered before definitive proof of infections and until the results of microbial investigation are available (de-escalation antimicrobial chemotherapy). On the other hand, several authors do not recommend glycopeptide administration in an attempt to limit nosocomial outbreaks of vancomycin-resistant enterococci (VRE) and staphylococci (VRS) and to avoid secondary drawbacks, such as nephrotoxicity and ototoxicity. De-escalation antimicrobial chemotherapy should be tailored to critically ill patients according to their clinical status, severity of illness and suspicion of sepsis or nosocomial pneumonia.