Adequate empiric antimicrobial therapy is crucial in terms of survival in patients with severe infections [1,2]. Moreover, inadequate empirical therapy significantly increases the length of hospitalization in critically ill patients with severe sepsis or septic shock . With these premises, when microbiological information is not available yet, the use of broad-spectrum antimicrobial(s) constitutes the backbone of the empirical therapy in critically ill patients.
Broad-spectrum antimicrobial treatment is defined as a combination of antibiotics which acts against a wide range of disease-causing bacteria. Certain families of antibiotics (i.e. piperacillin–tazobactam or carbapenems) pose a wide antimicrobial spectrum and their use as monotherapy is considered as broad-spectrum therapy. It is worth mentioning that carbapenems are the most commonly used antibiotics in the critical care setting, especially for nosocomial sepsis .
However, the use of broad-spectrum antimicrobial treatment is not without its drawbacks: antibiotic-related side effects, extra costs and the emergence of bacterial resistance. International guidelines recommend the use of broad-spectrum antibiotics to minimize the risk for inadequate antimicrobial treatment, which has been shown to increase mortality [5–7]. However, once the pathogen(s) are identified, the empiric antibiotic(s) should be stopped or reduced in number and/or narrowed in spectrum. This strategy termed ‘de-escalation therapy’ appears theoretically correct, capable of promoting therapeutic appropriateness and reducing costs.
HOW CAN WE DEFINE DE-ESCALATION THERAPY?
Regrettably, a precise consensus definition of de-escalation is still lacking. De-escalation generally refers to a reduction in the spectrum of administered antibiotics through the discontinuation of antibiotics providing activity against nonpathogenic organisms, discontinuation of antibiotics with similar activity or switching to an agent with narrower spectrum. De-escalation is mostly accomplished by a reduction in the number of antibiotics prescribed [8,9]. Furthermore, the period of time during which the de-escalation should be performed has not been determined.
Madaras-Kelly et al. [10▪▪] have developed a numerical score to measure the microbial spectrum of antibiotic regimens using a modified Delphi method. This score could be a tool to quantify the rate of de-escalation, although its clinical applicability is uncertain.
Selective pressure exerted by broad-spectrum antibiotics plays a crucial role in the emergence of multidrug-resistant bacteria. The impact on the microbiota of some antibiotics remains for extended periods of time . The capability of the different antimicrobial agents in altering endogenous flora should be considered for the election of the directed therapy.
RATES OF DE-ESCALATION THERAPY IN CRITICALLY ILL PATIENTS
Roughly, this strategy is accomplished in approximately 35–50% of the patients with severe sepsis [8,9,12▪]. The de-escalation rate in ventilator-associated pneumonia (VAP) is very similar, although this strategy is performed only in less than 10% when multidrug-resistant pathogens are implicated [13–15]. However, comparisons among studies are difficult by the absence of standardized definitions of de-escalation, variability of the empirical regimens and the differences in patient populations.
Multiple reasons could explain these low rates of antibiotic de-escalation in the critical care setting: reluctance to change an antibiotic regimen that has proven to be effective, lack of microbiological data, poor understanding of how to de-escalate and the controversial data about its effectiveness and safety.
CURRENT EVIDENCE ABOUT EFFECTIVENESS AND SAFETY OF DE-ESCALATION THERAPY
Observational studies and randomized clinical trials have been conducted to evaluate the clinical impact and safety of de-escalation therapy in patients with severe infections and sepsis.
De-escalating strategies have been evaluated particularly in VAP, in which the potential risk of multidrug-resistant microorganisms is relatively high. Several studies have shown that de-escalation therapy can be safely provided to patients with ICU-acquired pneumonia and is even associated with lower mortality. Rello et al.  found that the mortality rate of VAP patients de-escalated was significantly lower than in patients with no modification of the empiric antibiotics (18.4 vs. 43.4%). Similarly, Kollef et al. [16▪] analysed 398 patients with VAP. The mortality rate was significantly lower among patients in whom therapy was de-escalated (17.0%), than in those experiencing therapy escalation (42.6%) and those in whom empirical regimen was not modified (23.7%; P = 0.001) [16▪]. Giantsou et al. , in a prospective observational study of VAP patients, found that antibiotic de-escalation was associated with significantly decreased 28-day mortality (12 vs. 43.5%). In another retrospective study, the de-escalation group had a significantly lower pneumonia-related 30-day mortality rate than in the group with fixed therapy (2.3 vs. 14%) . Of note, a multivariate analysis was not performed in these studies to assess whether this strategy was a protective factor for mortality. De-escalation therapy was also associated with shorter length of stay and with a reduction of resource utilization .
Alvarez-Lerma et al. , in a prospective multicentre study of patients with VAP, reported that the ICU length of stay was significantly longer in culture-positive patients with narrower-spectrum alternatives who were not de-escalated than in culture-positive patients whose therapy was modified accordingly (36.7 vs. 23.7 days). In this study, de-escalation therapy was not a variable independently associated with mortality by logistic regression analysis.
This antibiotic policy has also been assessed in patients with severe sepsis. Morel et al.  evaluated a heterogeneous group of critically ill patients with severe infections. Their main finding was that de-escalation therapy was associated with a significant reduction of recurrent infection (19 vs. 5%, P = 0.01) without changes in mortality .
To evaluate the effect of antibiotic de-escalation on outcomes, we conducted a prospective, observational study involving adults admitted to the ICU with severe sepsis or septic shock. De-escalation of the initial regimen was performed in 219 patients (35%), more commonly in medical than in surgical patients. The hospital mortality rate was 27% in patients with therapy de-escalation, 33% in those with no treatment change and 43% in those with treatment escalation (P = 0.006). Propensity score adjusted multivariate regression analysis identified de-escalation therapy to be as protective for hospital 90-day mortality in the entire cohort, as well as in patients who had received empirical adequate therapy [12▪].
In 754 patients with bacteremic sepsis, Koupetori et al.  evaluated the impact on the outcome of de-escalation therapy in two periods. Antibiotic streamlining did not impact final outcome, although this strategy led to survival benefit in the second study period . It is worth mentioning a recent observational study performed in 101 neutropenic patients with severe sepsis. In this selected population, de-escalation did not negatively affect any prognostic index including early (30-day) or late (1-year) mortality .
Randomized clinical trials
Two randomized clinical trials have evaluated this antimicrobial strategy with less positive results. Kim et al. [21▪] compared the use of imipenem as well as vancomycin and subsequent de-escalation with maintenance of the above-mentioned antimicrobial regimen in critically ill patients with hospital-acquired pneumonia (HAP). Mortality and length of stay were similar in both arms, but the emergence of multidrug-resistant organisms, especially methicillin-resistant Staphylococcus aureus (MRSA), was more frequent in the de-escalation group.
In patients with severe sepsis [22▪▪], a clinical trial concluded that, as compared with the continuation of the empirical treatment, de-escalation of antibiotics resulted in prolonged duration of ICU stay (primary outcome measure). A new infection occurred in 16 (27%) patients in the de-escalation group and six (11%) patients in the continuation group (P = 0.03). Again, mortality rate was unaffected.
HOW CAN DE-ESCALATION BE DONE?
De-escalation strategies may differ in different aspects. However, several steps should be always present (Fig. 1). First, before starting antimicrobial therapy, it is indispensable to obtain appropriate cultures in order to identify the pathogens responsible for septic conditions. This is the first step to carry out the de-escalation of the empirical therapy because it is challenging to discontinue empirical antibiotics without microbiological documentation of the infection.
Second, empirical antibiotic must cover all likely pathogens administering broad-spectrum antimicrobial therapy. Several studies have identified the adequacy of initial antibiotic therapy as an independent factor associated with de-escalation. As expected, the presence of multidrug-resistant bacteria hampered de-escalation [9,13,23]. Third, once the culprit pathogen(s) are identified, the empirical regimen should be adapted to the microbiological results.
De-escalation may also include switching antibiotics from intravenous to oral route. This change is rarely performed in critically ill patients but is a suitable option in certain situations.
De-escalation in infections caused by Gram-positive bacteria
Stopping antimicrobials directed against resistant Gram-positive bacteria (i.e. MRSA or Enterococcus faecium) should be the rule when these pathogens are not isolated in clinical samples. Switching to an agent with a narrower spectrum is easily performed for coverage of Gram-positive bacteria susceptible to beta-lactam antibiotics. There is compelling evidence that mortality and morbidity of severe infections caused by methicillin-susceptible S. aureus (MSSA) is significantly higher with a glycopeptide than with cloxacillin . Similarly, 30-day mortality of MSSA bacteremia is significantly lower with a beta-lactam with a high activity against MSSA (cloxacillin or cefazoline) than with a third-generation cephalosporin or with piperacillin–tazobactam . Very recently, glycopeptide use was associated with increased mortality in Enterococcus faecalis bacteremia [26▪].
De-escalation in infections caused by Gram-negative bacteria
Although the use of two active antimicrobials against certain Gram-negative bacilli (i.e. Pseudomonas aerugionsa) has been traditionally advocated , recent studies clearly demonstrated that the use of combination therapy in the directed therapy does not reduce recurrence rates and is not associated with any survival benefits [28,29]. Similarly, in severe infections caused by Acinetobacter baumannii, the use of monotherapy appears feasible without any obvious negative clinical impact . Therefore, if two active antimicrobials have been initiated to cover Gram-negative bacilli (GNB), switch to monotherapy can be safely performed once susceptibility results are available. Infections caused by carbapenemase-producing Klebsiella pneumoniae constitute an exception to this recommendation of monotherapy in the directed therapy .
More troublesome is usually the spectrum reduction of the empirically administered antibiotics for Gram-negative bacilli. Whenever possible, carbapenems must be stopped and switched to another antimicrobial with a narrower spectrum and less impact on resistance development . Carbapenems should be reserved because they frequently constitute the only therapeutic option against GNB such as extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae, Pseudomonas aeruginosa or A. baumannii. The likely alternatives are summarized in Table 1. Piperacillin–tazobactam is an option frequently used in the directed therapy when meropenem was used empirically . Recent studies have reported that the susceptibility of P. aeruginosa to imipenem improved after the introduction of ertapenem into hospital formularies .
De-escalation of antifungal therapy
Therapy with an antifungal agent should be discontinued if fungi (i.e. Candida spp.) are not present in clinical samples. Regarding the switch to an antifungal agent with narrower spectrum, echinocandins are considered the agents of choice for empirical treatment of critically ill patients with invasive candidiasis. In patients with candidemia, therapy until 14 days after the first negative blood culture is recommended. Recent guidelines include the recommendation to step-down to oral fluconazole after 10 days of intravenous treatment if the patient is clinically stable and blood cultures have become negative . The appropriate timing of antifungal step-down remains unclear, but this change might be done early and the use of intravenous fluconazole is a valid alternative provided that the strain is susceptible to the azole.
De-escalation therapy in culture-negative infections
Regarding de-escalation therapy, particularly challenging are the infections with negative cultures. In a retrospective study, antibiotic therapy was de-escalated in 75% of the patients with culture-negative HAP . Conversely, de-escalation was not performed in patients without a known pathogen in a series of critically ill patients with VAP . It is reasonable to cease empirical vancomycin in patients with culture-negative pneumonia and with no MRSA colonization . In other patients without microbiological documentation, streamlining of the empirical therapy should be left to clinical judgement as long as the patient has a good clinical course.
Antibiotic de-escalation is a well tolerated and highly recommended approach in critically ill patients with severe sepsis. The available data suggest that outcomes can be improved with its use. These outcomes include less antibiotic use, shorter duration of therapy and reduced mortality. Nevertheless, we also need more information about which patients can have therapy stopped altogether if cultures are negative and about the timing to carry out the streamlining of the empirical therapy.
Financial support and sponsorship
No financial support was received for this study.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar A, et al. Impact of adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive care unit with sepsis. Crit Care Med 2003; 31:2742–2751.
2. Vallés J, Rello J, Ochagavía A, et al. Community-acquired bloodstream infection in critically ill adult patients: impact of shock and inappropriate antibiotic therapy on survival
. Chest 2003; 123:1615–1624.
3. Garnacho-Montero J, Ortiz-Leyba C, Herrera-Melero I, et al. Mortality and morbidity attributable to inadequate empirical antimicrobial therapy in patients admitted to the ICU with sepsis: a matched cohort study. J Antimicrob Chemother 2008; 61:436–441.
4. Díaz-Martín A, Martínez-González ML, Ferrer R, et al. Edusepsis Study Group. Antibiotic prescription patterns in the empiric therapy of severe sepsis: combination of antimicrobials with different mechanisms of action reduces mortality. Crit Care 2012; 16:R223.
5. American Thoracic Society/Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare associated pneumonia. Am J Respir Crit Care Med 2005; 171:388–416.
6. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock: 2012. Crit Care Med 2013; 41:580–637.
7. Averbuch D, Orasch C, Cordonnier C, et al. ECIL4, a joint venture of EBMT, EORTC, ICHS, ESGICH/ESCMID and ELN. European guidelines for empirical antibacterial therapy for febrile neutropenic patients in the era of growing resistance: summary of the 2011 4th European Conference on Infections in Leukemia. Haematologica 2013; 98:1826–1835.
8. Morel J, Casoetto J, Jospé R, et al. De-escalation as part of a global strategy of empiric antibiotherapy management. A retrospective study in a medico-surgical intensive care unit. Crit Care 2010; 14:R225.
9. Heenen S, Jacobs F, Vincent J-L. Antibiotic strategies in severe nosocomial sepsis: why do we not deescalate more often? Crit Care Med 2012; 40:1404–1409.
10▪▪. Madaras-Kelly K, Jones M, Remington R, et al. Development of an antibiotic spectrum score based on Veterans Affairs culture and susceptibility data for the purpose of measuring antibiotic de-escalation: a modified Delphi approach. Infect Control Hosp Epidemiol 2014; 35:1103–1113.
An interesting initiative to develop a numerical score to measure the microbial spectrum of antibiotic regimens.
11. Jernberg C, Löfmark S, Edlund C, Jansson JK. Long-term impacts of antibiotic exposure on the human intestinal microbiota. Microbiology 2010; 156:3216–3223.
12▪. Garnacho-Montero J, Gutiérrez-Pizarraya A, Escoresca-Ortega A, et al. De-escalation of empirical therapy is associated with lower mortality in patients with severe sepsis and septic shock. Intensive Care Med 2014; 40:32–40.
A prospective, observational study demonstrating that de-escalation of antibiotic therapy in patients with sepsis was associated with significantly decreased hospital and 90-day mortality.
13. Rello J, Vidaur L, Sandiumenge A, et al. De-escalation therapy in ventilator-associated pneumonia. Crit Care Med 2004; 32:2183–2190.
14. Eachempati SR, Hydo LJ, Shou J, Barie PS. Does de-escalation of antibiotic therapy for ventilator-associated pneumonia affect the likelihood of recurrent pneumonia or mortality in critically ill surgical patients? J Trauma 2009; 66:1343–1348.
15. Joung MK, Lee JA, Moon SY, et al. Impact of de-escalation therapy on clinical outcomes for intensive care unit-acquired pneumonia. Crit Care 2011; 15:R79.
16▪. Kollef MH, Morrow LE, Niederman MS, et al. Clinical characteristics and treatment patterns among patients with ventilator-associated pneumonia. Chest 2006; 129:1210–1218.
A multicentre experience with VAP therapy documenting the frequency and consequences of a de-escalation strategy. The use of a de-escalation approach led to reduced mortality compared with an escalating therapy strategy or with no change.
17. Giantsou E, Liratzopoulos N, Efraimidou E, et al. De-escalation therapy rates are significantly higher by bronchoalveolar lavage than by tracheal aspirate. Intensive Care Med 2007; 33:1533–1540.
18. Alvarez-Lerma F, Alvarez B, Luque P, et al. ADANN Study Group. Empiric broad-spectrum antibiotic therapy of nosocomial pneumonia in the intensive care unit: a prospective observational study. Crit Care 2006; 10:R78.
19. Koupetori M, Retsas T, Antonakos N, et al. Hellenic Sepsis Study Group. Bloodstream infections and sepsis in Greece: over-time change of epidemiology and impact of de-escalation on final outcome. BMC Infect Dis 2014; 14:272.
20. Mokart D, Slehofer G, Lambert J, et al. De-escalation of antimicrobial treatment in neutropenic patients with severe sepsis: results from an observational study. Intensive Care Med 2014; 40:41–49.
21▪. Kim JW, Chung J, Choi SH, et al. Early use of imipenem/cilastatin and vancomycin followed by de-escalation versus conventional antimicrobials without de-escalation for patients with hospital-acquired pneumonia in a medical ICU: a randomized clinical trial. Crit Care 2012; 16:R28.
A unicentre, open-label randomized trial of patients with HAP who were assigned to de-escalation or continuation of empirical regimen. Although mortality was similar in both groups, emergence of multidrug resistance was significantly higher in the de-escalation group.
22▪▪. Leone M, Bechis C, Baumstarck K, et al. AZUREA Network Investigators. De-escalation versus continuation of empirical antimicrobial treatment in severe sepsis: a multicenter non-blinded randomized noninferiority trial. Intensive Care Med. 2014; 40:1399–1408.
A multicentre, non-blinded, randomized, noninferiority trial of patients with severe sepsis who were assigned to de-escalation or continuation of empirical antimicrobial treatment. Although mortality was similar in both groups, length of ICU stay was significantly longer in the de-escalation group.
23. Gonzalez L, Cravoisy A, Barraud D, et al. Factors influencing the implementation of antibiotic de-escalation and impact of this strategy in critically ill patients. Crit Care 2013; 17:R140.
24. González C, Rubio M, Romero-Vivas J, et al. Bacteremic pneumonia due to Staphylococcus aureus: a comparison of disease caused by methicillin-resistant and methicillin-susceptible organisms. Clin Infect Dis 1999; 29:1171–1177.
25. Paul M, Zemer-Wassercug N, Talker O, et al. Are all beta-lactams similarly effective in the treatment of methicillin-sensitive Staphylococcus aureus bacteraemia? Clin Microbiol Infect 2011; 17:1581–1586.
26▪. Foo H, Chater M, Maley M, van Hal SJ. Glycopeptide use is associated with increased mortality in Enterococcus faecalis bacteraemia. J Antimicrob Chemother 2014; 69:2252–2257.
A recent study demonstrating that compared with the use of a beta-lactam, glycopeptide treatment was an independent predictor of 30-day mortality in patients with E. faecalis bacteremia.
27. Safdar N, Handelsman J, Maki DG. Does combination antimicrobial therapy reduce mortality in Gram-negative bacteraemia? A meta-analysis. Lancet Infect Dis 2004; 4:519–527.
28. Garnacho-Montero J, Sa-Borges M, Sole-Violan J, et al. Optimal management therapy for Pseudomonas aeruginosa
ventilator-associated pneumonia: an observational, multicenter study comparing monotherapy with combination antibiotic therapy. Crit Care Med 2007; 35:1888–1895.
29. Peña C, Suarez C, Ocampo-Sosa A, et al. Spanish Network for Research in Infectious Diseases (REIPI). Effect of adequate single-drug vs combination antimicrobial therapy on mortality in Pseudomonas aeruginosa bloodstream infections: a post hoc analysis of a prospective cohort. Clin Infect Dis 2013; 57:208–216.
30. López-Cortés LE, Cisneros JM, Fernández-Cuenca F, et al. GEIH/REIPI-Ab2010 Group. Monotherapy versus combination therapy for sepsis due to multidrug-resistant Acinetobacter baumannii
: analysis of a multicentre prospective cohort. J Antimicrob Chemother 2014; 69:3119–3126.
31. Tumbarello M, Viale P, Viscoli C, et al. Predictors of mortality in bloodstream infections caused by Klebsiella pneumoniae carbapenemase-producing K. pneumoniae: importance of combination therapy. Clin Infect Dis 2012; 55:943–950.
32. Kuo HY, Chang KC, Kuo JW, et al. Imipenem: a potent inducer of multidrug resistance in Acinetobacter baumannii. Int J Antimicrob Agents 2012; 39:33–38.
33. De Waele JJ, Ravyts M, Depuydt P, et al. De-escalation after empirical meropenem treatment in the intensive care unit: fiction or reality? J Crit Care 2010; 25:641–646.
34. Goldstein EJ, Citron DM, Peraino V, et al. Introduction of ertapenem into a hospital formulary: effect on antimicrobial usage and improved in vitro susceptibility of Pseudomonas aeruginosa. Antimicrob Agents Chemother 2009; 53:5122.
35. Cornely OA, Bassetti M, Calandra T, et al. ESCMID Fungal Infection Study Group. ESCMID* guideline for the diagnosis and management of Candida diseases 2012: non-neutropenic adult patients. Clin Microbiol Infect 2012; 18 (Suppl 7):19–37.
36. Schlueter M, James C, Dominguez A, et al. Practice patterns for antibiotic de-escalation in culture-negative healthcare-associated pneumonia. Infection 2010; 38:357–362.
37. Boyce JM, Pop OF, Abreu-Lanfranco O, et al. A trial of discontinuation of empiric vancomycin therapy in patients with suspected methicillin-resistant Staphylococcus aureus health care-associated pneumonia. Antimicrob Agents Chemother 2013; 57:1163–1168.