Rational antimicrobial treatment is a key component in the management of critically ill patients. At the 2016 Euroanaesthesia Conference (London, UK), the European Society of Anaesthesia Intensive Care Scientific Subcommittee convened an expert panel on antibiotic therapy. A summary of the panel's conclusions is provided in this narrative review.
Balancing the risk and the benefit
Routine screening for infection is not usually justified in critically ill patients because the likelihood of a positive culture is highly dependent on the clinical context, which is very nuanced in the ICU setting.1 The timing of empirical antibiotic administration and its association with better survival is one of the best confirmations in clinical practice.2 Once a decision has been made to administer empirical antimicrobial therapy, one should also determine which pathogens are targeted and tailor antimicrobial dose and spectrum, based on the individual patient's clinical characteristics. The stakes are exceedingly high for these decisions as the risk of death increases with initiation of inadequate antimicrobial therapy.3
In clinical scenarios involving preemptive treatment or antibiotic prophylaxis, even when the appropriate antimicrobial has been selected, both timing4 and mode of delivery5 may affect the patients’ overall outcome. However, there is a striking paucity of literature on this topic in the critically ill patient group (e.g. trauma, urgent colorectal surgery and multiorgan failure). Studies that use selection pressure from preemptive broad-spectrum treatment in at-risk surgical patients may be associated with a drift towards resistant species which raises additional concerns.6
Even more confounding is the fact that although standard dosing is practised in most places, antimicrobial therapy may be more effective when individualised. Variables such as the microbiome of the critically ill patient,7 predisposing host genetic susceptibility and differences in gene expression in reaction to infection, as well as the characteristics of the infecting organism and its susceptibility, all lead to a highly unpredictable response to therapy.8,9 Sepsis itself is also accompanied by altered drug pharmacokinetics and pharmacodynamics. Decreases in serum albumin concentrations and redistribution of fluids in the various body compartments may increase antimicrobial volume of distribution and clearance, leading to insufficient drug blood levels (particularly of time-dependent antimicrobials). Alternatively, impaired kidney function may be accompanied by antimicrobial overdose unless the patient is requiring continuous renal replacement therapy. In this case, residual diuresis and type of membrane used might represent important factors for adjusting drug doses, especially relevant when administering beta-lactams.10–12
Too much of a good thing?
The need to ensure that ICU patients with confirmed bacterial infections receive timely and appropriate antimicrobial coverage often causes an initial excess of treatment. Thus, about half of all empirical antibiotics ordered in ICU patients are continued for at least 72 h despite a specific confirmed infection.13 Clinical reassessment and information regarding culture growth should provide the information required to determine whether antimicrobial therapy might be de-escalated by either narrowing the spectrum or discontinuing altogether. However, cultures do not always yield the required information and clinical findings are often misleading. Significantly, adherence to guidelines,14,15 including de-escalation and avoidance of antibiotic over-treatment,16,17 has been associated with improved outcomes. This shows that current guidelines together with a daily clinical patient assessment need to be better integrated as a multidisciplinary approach in ICU settings.18
Shortening the duration of antimicrobial therapy is an additional powerful strategy for reducing antimicrobial resistance. The duration of therapy is well defined in most guidelines, providing ample opportunity for individual judgment and preferences. At the same time, clinicians are often loath to curtail therapy, particularly when the infection may be life threatening. As a result, antimicrobial therapy is often extended under the pretext of tailoring to the clinical course of the patient, and many clinicians uphold the use of serial blood testing for the presence of various inflammatory biomarkers as decision support tools (e.g. procalcitonin).19,20
Empirical antibiotic treatment for severe infections
Patients admitted to the ICU are increasingly more fragile – they are older and suffer from more comorbidities than patients admitted to the ICU a decade ago.21 In parallel, there is an increase in multidrug resistant organisms that are no longer susceptible to first-line antibiotic therapy [e.g. methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococci species and extended-spectrum beta-lactamase-producing organisms]. The combination of the aforementioned aspects creates a vicious cycle of escalating antimicrobial treatment.22 Intensive care physicians naturally prioritise the good of their patient over epidemiological considerations of antibiotic resistance.23 However, we must never forget that the way we use antibiotics today will have consequences for our patients tomorrow.
Patients with septic shock who receive inappropriate antibiotic therapy have an excess mortality of 30 to 40% compared with those without shock.24 Observational studies also have shown a reduction in mortality from the first hour of antibiotic administration in patients with severe sepsis and septic shock.2,25 Conversely, some studies have also shown that delaying antimicrobial therapy until information regarding the pathogen emerges does not worsen mortality and is associated with reduced antimicrobial use.26 These contradictory results probably stem from differences in severity of illness and case mix. Appropriate patient selection for each treatment option determines treatment success.
Use and misuse of antibiotic combinations
Administration of two antibiotics or more at the same time is defined as an ‘antibiotic combination’. This definition takes into consideration neither the spectrum of antibacterial coverage nor the fact that two molecules may be included in one drug. Clinicians opt for antibiotic combinations for several reasons: to broaden the spectrum of treatment (most commonly), to achieve a synergistic effect or to reduce the emergence of resistant bacterial strains.27 Treatment goals may differ in empirical vs. targeted antibiotic combinations (Fig. 1).
Use of antibiotic combinations for Empirical treatment
During empirical treatment, antibiotic combinations are mostly used to broaden the spectrum of antibacterial coverage. For instance, during treatment of sepsis and septic shock, an antibiotic active against MRSA (i.e. vancomycin or linezolid or daptomycin) may be added to a beta-lactam active against Pseudomonas aeruginosa [e.g. piperacillin–tazobactam, ceftazidime, cefepime or a carbapenem (excluding ertapenem)] chosen in accordance with local epidemiology.28 In most cases, an aminoglycoside (e.g. gentamicin and amikacin) or a fluoroquinolone (ciprofloxacin and levofloxacin) is added to a beta-lactam. Of note, there are no randomised clinical trials supporting any antibiotic combination, although this is common practice, and it might be especially relevant in the event of a bacteraemic episode.28
The utility of antibiotic combinations varies in different clinical setting.29 Thus in patients with community-acquired pneumonia, for example, adding a macrolide (such as erythromycin or clarithromycin) to a beta-lactam has often been advocated because of the anti-inflammatory effects of macrolides.30 Martin-Loeches et al.,31 in a prospective multicentre study of patients with severe community-acquired pneumonia, found lower ICU mortality in patients treated with macrolides than in those treated with fluoroquinolones. Restrepo et al.32 also found that treatment with macrolides was associated with decreased mortality in patients with severe sepsis due to community-acquired pneumonia, even when a macrolide-resistant pathogen was isolated. Conversely, a recent randomised clinical trial conducted in the Netherlands showed that antibiotic monotherapy was as efficient as combination antibiotic therapy for treating patients with community-acquired pneumonia admitted to the ward.33 A meta-analysis34 showed that antibiotic monotherapy is associated with fewer recurrent infections than antibiotic combinations in patients with sepsis. Similar meta-analytical findings have been reported in patients with febrile neutropenia, suggesting that even during some forms of immune suppression, antibiotic combination therapy may not be required.35 In contrast, the meta-analysis by Kumar et al., 36 which included only patients with severe sepsis and septic shock, found that antibiotic combination therapy provided survival benefit, but only in high-risk patients.26 This last analysis highlighted that inconsistency in existing studies, in both inclusion criteria and outcomes, probably confounded the conclusions.26
Therefore, randomised controlled trials have thus far failed to support the common conviction that antibiotic combinations are associated with a reduced likelihood of failed pathogen coverage during empirical treatment (which is important in an era of emerging multidrug-resistant organisms). However, selective inclusion criteria and inconsistent endpoints may have masked this effect. At the moment, there is insufficient evidence to favour one form of treatment over another. When there is a risk of infection with multidrug-resistant organisms in a critically ill patient with complex pathology, one must consider antibiotic combinations for empirical treatment.37
Use of antibiotic combinations for targeted treatment
In the presence of specific infections, antibiotic combinations are usually given to achieve a synergistic effect. Synergy is defined as a more than 2-log increase in bactericidal activity in vitro, as compared with the bactericidal activity of each agent alone (Fig. 1). In-vitro models of infection demonstrate that adding an aminoglycoside (e.g. gentamicin) to a beta-lactam (e.g. ampicillin)38 confers a synergistic effect. However, there is little proof that laboratory findings of synergy translate into clinical benefit. For example, combination therapy is often suggested in the hope of achieving a synergistic effect against P. aeruginosa or Acinetobacter baumannii in ICU patients. However, studies show that the combination of fluoroquinolones and beta-lactams provides only sporadic synergy,27 whereas fluoroquinolones and aminoglycosides provide no synergy at all.39 Meta-analyses have also failed to show clinical advantage with antibiotic combinations in patients infected with P. aeruginosa or A. baumannii.40,41 Based on these data, the effectiveness of administering antibiotic combination therapy when treating these pathogens seems very overrated. So the question arises as to where antibiotic combinations still have a place? The answer may be in the treatment of infections caused by pandrug-resistant Acinetobacter species42 and when infection involves non-native material. However, although, two meta-analyses have failed to demonstrate benefit with the use of antibiotic combination therapy 34,43 such therapy is still recommended in patients with proven MRSA endocarditis on a prosthetic valve.44
A note of caution regarding administration of antibiotic combinations with the intent of achieving a synergistic effect: some antibiotic combinations may actually be antagonistic. Classic examples are the combination of beta-lactams with chloramphenicol or rifampicin with vancomycin. Furthermore, antibiotic combination therapy may generate more adverse events than monotherapy. Among the adverse events described in the literature are nephrotoxicity associated with the use of aminoglycosides and polymixins42 and secondary Clostridium difficile infection.45
In summary, empirical use of combined antibiotic therapy is not based upon compelling evidence. However, if the risk of infection due to multidrug-resistant organisms is significant, it seems reasonable to combine antibiotics to cover a greater spectrum of resistance. Under these circumstances, early and repeated reassessment of possible de-escalation is mandatory. For targeted treatment, antibiotic combinations should only be used when there is documented evidence of infection due to pandrug-resistant pathogens or when there is evidence of infection in specific sites involving non-native material.
Clinicians at the forefront of intensive care understandably tend to be more concerned with the likelihood of patient deterioration as a result of insufficient/inappropriate antibiotic therapy than with the long-term implications of antibiotic resistance even in the individual patient. To date, not all ICUs follow antimicrobial prescribing guidelines, and even when such have been instituted, adherence rates are 40 to 60% at best.31 However, guidelines alone do not necessarily change antimicrobial consumption. Multifaceted antibiotic stewardship is a better approach towards decreasing unnecessary antibiotic treatment in the ICU, and antibiotic usage can also be improved through educational interventions.46 Antibiotic stewardship programs have an important role in reminding the clinician that leucocytosis and fever do not always indicate the presence of infection13 or treatment failure.47 At times, it is easier for a physician less involved in patient care to remind those directly involved that one is allowed to use clinical judgement and to focus treatment on the patient as a whole, rather than on the results of laboratory cultures. An additional important role of antibiotic stewardship is promotion of de-escalation. De-escalation of antibiotic therapy is well tolerated and feasible in ICU patients provided there is adequate interpretation of culture results.17,23 Interactive discussion between the treating ICU team and the in-hospital consulting microbiology team is crucial to ensure optimisation of antibiotic therapy. Although microbiological results should be interpreted within their clinical context (e.g. severity of the patient's illness and risk factors), local epidemiology and microbiological information [including sensitivity and minimum inhibitory concentrations (MICs)] are both required for optimal drug choice and dosing. The MIC is defined as the lowest concentration of an antimicrobial that will inhibit visible growth of a microorganism. Bacterial strains with low MICs are easier to eradicate than those with high ones, and this information, which is often available only to the microbiologist, is of great importance when deciding what drug and dose to choose. Multidisciplinary discussion is therefore imperative when making decisions regarding antimicrobial care in critically ill patients. In the future, rapid diagnostic tests are expected to change the landscape of clinical management of infections, allowing earlier targeted decision-making. Source control and total drug monitoring are currently being assessed as quality measures indicating the adequacy of clinical practice. Should these be adopted, the decision-making process of all clinicians involved in ICU patient care will be under surveillance. Thus, all those involved must be aware of the protocols for antibiotic management in ICU (Fig. 2).
What to give and when to stop
At the time of disease onset or ICU admission, the empirical antibiotic therapy prescribed should be broad enough to cover all likely pathogens when taking into consideration both local strains and/or the suspected source of infection. In septic patients, this usually entails a combination of a broad-spectrum beta-lactam active against Pseudomonas species (e.g. imipenem, meropenem, piperacilin-tazobactam, ceftazidime, and cefepime) and either an aminoglycoside (e.g. gentamicin, amikacin, and tobramycin). Currently where patients have high levels of resistance, carbapenems and fluoroquinolone should be avoided as a first-line therapy, if possible, and considered only with specific risk factors such as known resistance or previous antibiotic use. Additional coverage for MRSA (e.g. vancomycin or linezolid when the lung is the suspected source of infection, or vancomycin or daptomycin if other sources are suspected) depends on local epidemiology. Such additional treatment is appropriate in areas with a high prevalence of MRSA. Every ICU should develop protocols for empirical antibiotic therapy based on local epidemiology. For early-onset nosocomial infections in patients admitted to ICU settings with low rates of multidrug-resistant organisms, or without risk factors for such, and who are not shocked, a narrow spectrum prescription based on third generation cephalosporin (cefotaxime or ceftriaxone) alone may be selected.48 For late-onset infections and for patients at risk of infection with multidrug-resistant organisms, or in clinical settings with a high prevalence of multidrug-resistant organisms, a broader spectrum beta-lactam (active against nonfermentative gram-negative bacilli) in combination with either an aminoglycoside or a fluoroquinolone, with or without coverage for MRSA dependent on local epidemiology, might be a reasonable option.
To avoid the development of drug resistant bacteria prolonged and indiscriminate use of broad-spectrum antibiotics should be avoided. De-escalation should always be considered on days 3 to 5, or whenever microbiological results are available. An antibiotic with an appropriate narrower spectrum of activity (usually a beta-lactam) should be substituted for the initial broad-spectrum antibiotic combination. If infection with extended spectrum beta-lactamase-producing organisms is suspected, a carbapenem should be considered initially for empirical antibiotic therapy. However piperacillin/tazobactam could also be suitable for confirmed, non-severe, extended spectrum beta-lactamase producing organism infections (especially with Escherichia coli) with low MICs. Limiting the duration of antibiotic therapy will limit the emergence of antibiotic resistance without compromising patient prognosis. Most severe infections are eradicated within 7 to 10 days of antibiotic therapy. Even shorter durations (e.g. 5 to 7 days) can be recommended for intra-abdominal infections in patients with low severity who are not critically ill, in conjunction with adequate source control.49 Adapting the duration of treatment based on biomarker kinetics is an emerging option.50
Conclusion: lessons from the past, tracks for the future
Currently, along with adequate source control, antibiotics remain the ICU physician's most powerful weapon against infection. However, a worldwide crisis of rapidly emerging antibiotic-resistant pathogens comprises a very real and current threat. Resistant bacterial strains already threaten our ability to provide adequate antibiotic coverage for critically ill infected patients today, and such resistant strains are expected to become ever more prevalent in the future. Attaining the precarious balance between efficacious patient care and preservation of the existing delicate ecological balance is no mean feat. All physicians working with critically ill patients should be aware of the double role they assume when determining the optimal therapy for their patient: while striving to save the life of their current patient, for the sake of future patients they should not compromise the local bacterial ecology. Worldwide, physicians involved in the care of critically ill patients have a major responsibility in this team effort to minimise the emergence of antibiotic-resistant pathogens. Only in this way can we make well informed decisions that allow us to treasure last-line antibiotics (a rapidly diminishing resource) for the sake of future generations of ICU physicians and their patients. One should always keep in mind that, beyond the collective issues, unnecessary antibiotic overuse might be harmful to the individual patient over the space of a few days (e.g. emergence of multidrug resistant organisms, C. difficile infection); thus, antimicrobial resistance is both an individual and collective challenge.
Acknowledgements relating to this article
Assistance with the special article: the authors would like to thank Zieta O’Hagan for English language editing.
Financial support and sponsorship: none.
Conflicts of interest: the authors have no conflict of interest in relation to the content and recommendations provided in this manuscript.
Presentation: Euroanaesthesia congress 2016, London.
1. Warren MM, Gibb AP, Walsh TS. Antibiotic prescription practice in an intensive care unit using twice-weekly collection of screening specimens: a prospective audit in a large UK teaching hospital. J Hosp Infect
2. Ferrer R, Martin-Loeches I, Phillips G, et al. Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: results from a guideline-based performance improvement program. Crit Care Med
2014; 42:1749–1755. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24717459
. [Cited 13 Oct 2014; Internet].
3. Díaz-Martín A, Martínez-González ML, Ferrer R, et al. Antibiotic prescription patterns in the empiric therapy of severe sepsis: combination of antimicrobials with different mechanisms of action reduces mortality. Crit Care
2012; 16:R223Available from: http://www.ncbi.nlm.nih.gov/pubmed/23158399
. [Cited 30 Jan 2013; Internet].
4. Mackeen AD, Packard RE, Ota E, Berghella V, Baxter JK. Timing of intravenous prophylactic antibiotics for preventing postpartum infectious morbidity in women undergoing cesarean delivery. Cochrane Database Syst Rev
2014; CD009516doi: 10.1002/14651858.CD009516.pub2.
5. Liberati A, D’Amico R, Pifferi S, Torri V, Brazzi L, Parmelli E. Antibiotic prophylaxis to reduce respiratory tract infections and mortality in adults receiving intensive care. Cochrane Database Syst Rev
6. Bouza E, Granda MJP, Hortal J, et al. Preemptive broad-spectrum treatment for ventilator-associated pneumonia in high-risk patients. Intensive Care Med
2013; 39:1547–1555. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23812339
. [Cited 31 Jul 2013; Internet].
7. Dickson RP. The microbiome and critical illness. Lancet Respir Med
8. Goncalves-Pereira J, Povoa P. Antibiotics in critically ill patients: a systematic review of the pharmacokinetics of beta-lactams. Crit Care
9. Ulldemolins M, Vaquer S, Llaurado-Serra M, et al. Beta-lactam dosing in critically ill patients with septic shock and continuous renal replacement therapy. Crit Care
10. Ulldemolins M, Martin-Loeches I, Llaurado-Serra M, et al. Piperacillin population pharmacokinetics in critically ill patients with multiple organ dysfunction syndrome receiving continuous venovenous haemodiafiltration: effect of type of dialysis membrane on dosing requirements. J Antimicrob Chemother
11. Ulldemolins M, Soy D, Llaurado-Serra M, et al. Meropenem population pharmacokinetics in critically ill patients with septic shock and continuous renal replacement therapy: influence of residual diuresis on dose requirements. Antimicrob Agents Chemother
12. Ulldemolins M, Vaquer S, Llauradó-Serra M, et al. Beta lactam dosing in critically ill patient with septic shock and continuous renal replacement therapy. Crit Care
13. Martin-Loeches I, Levy MM, Artigas A. Management of severe sepsis: advances, challenges, and current status. Drug Des Devel Ther
2015; 9:2079–2088. Available from: http://www.ncbi.nlm.nih.gov/pubmed/25926718
15. Ibrahim EH, Ward S, Sherman G, et al. Experience with a clinical guideline for the treatment of ventilator-associated pneumonia. Crit Care Med
2001; 29:1109–1115. Available from: http://www.ncbi.nlm.nih.gov/pubmed/11395584
. [Cited 29 Mar 2014; Internet].
16. 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. Available from: http://www.ncbi.nlm.nih.gov/pubmed/24026297
. [Cited 9 Oct 2014; Internet].
17. Leone M, Bechis C, Baumstarck K, et al. De-escalation versus continuation of empirical antimicrobial treatment in severe sepsis: a multicenter nonblinded randomized noninferiority trial. Intensive Care Med
2014; 40:1399–1408. Available from: http://icmjournal.esicm.org/journal/134/0/10/3411/0/
18. Bilimoria KY, Sohn M-W, Chung JW, et al. Association between state medical malpractice environment and surgical quality and cost in the United States. Ann Surg
19. Woodhead M, Blasi F, Ewig S, et al. Guidelines for the management of adult lower respiratory tract infections. Eur Respir J
20. Kopterides P, Siempos II, Tsangaris I, et al. Procalcitonin-guided algorithms of antibiotic therapy in the intensive care unit: a systematic review and meta-analysis of randomized controlled trials. Crit Care Med
21. Bagshaw SM, Webb SAR, Delaney A, et al. Very old patients admitted to intensive care in Australia and New Zealand: a multicentre cohort analysis. Crit Care
22. Martin-Loeches I, Povoa P, Rodríguez A, et al. Incidence and prognosis of ventilator-associated tracheobronchitis (TAVeM): a multicentre, prospective, observational study. Lancet Respir Med
2015; 3:859–868. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26472037
23. Martin-Loeches I, Diaz E, Valles J. Risks for multidrug-resistant pathogens in the ICU. Curr Opin Crit Care
24. Vallés J, Palomar M, Alvárez-Lerma F, et al. Evolution over a 15-year period of clinical characteristics and outcomes of critically ill patients with community-acquired bacteremia. Crit Care Med
2013; 41:76–83. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23222266
. [Cited 30 Mar 2013; Internet].
25. Kumar AA, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med
2006; 34:1589–1596. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16625125
. [Cited 28 Feb 2013; Internet].
26. Hranjec T, Rosenberger LH, Swenson B, et al. Aggressive versus conservative initiation of antimicrobial treatment in critically ill surgical patients with suspected intensive-care-unit-acquired infection: a quasi-experimental, before and after observational cohort study. Lancet Infect Dis
2012; 12:774–780. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22951600
27. Tamma PD, Cosgrove SE, Maragakis LL. Combination therapy for treatment of infections with gram-negative bacteria. Clin Microbiol Rev
28. Heyland DK, Dodek P, Muscedere J, et al. Randomized trial of combination versus monotherapy for the empiric treatment of suspected ventilator-associated pneumonia. Crit Care Med
2008; 36:737–744. Available from: http://www.ncbi.nlm.nih.gov/pubmed/18091545
. [Cited 18 Mar 2013; Internet].
29. Martin-Loeches I, J Schultz M, Vincent J-L, et al. Increased incidence of co-infection in critically ill patients with influenza. Intensive Care Med
2017; 43:48–58. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27709265
30. Restrepo MI, Sole-Violan J, Martin-Loeches I. Macrolide therapy of pneumonia: is it necessary, and how does it help? Curr Opin Infect Dis
31. Martin-Loeches I, Lisboa T, Rodriguez A, et al. Combination antibiotic therapy with macrolides improves survival in intubated patients with community-acquired pneumonia. Intensive Care Med
2010; 36:612–620. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19953222
32. Restrepo MI, Mortensen EM, Waterer GW, et al. Impact of macrolide therapy on mortality for patients with severe sepsis due to pneumonia. Eur Respir J
33. Postma DF, van Werkhoven CH, van Elden LJR, et al. Antibiotic treatment strategies for community-acquired pneumonia in adults. N Engl J Med
34. Paul M, Lador A, Grozinsky-Glasberg S, Leibovici L. Beta lactam antibiotic monotherapy versus beta lactam-aminoglycoside antibiotic combination therapy for sepsis. Cochrane Database Syst Rev
2014; CD003344doi: 10.1002/14651858.CD003344.pub3.
35. Furno P, Bucaneve G, Del Favero A. Monotherapy or aminoglycoside-containing combinations for empirical antibiotic treatment of febrile neutropenic patients: a meta-analysis. Lancet Infect Dis
36. Kumar A, Safdar N, Kethireddy S, Chateau D. A survival benefit of combination antibiotic therapy for serious infections associated with sepsis and septic shock is contingent only on the risk of death: a meta-analytic/meta-regression study. Crit Care Med
37. Micek ST, Welch EC, Khan J, et al. Empiric combination antibiotic therapy is associated with improved outcome against sepsis due to Gram-negative bacteria: a retrospective analysis. Antimicrob Agents Chemother
38. Ramos MC, Grayson ML, Eliopoulos GM, Bayer AS. Comparison of daptomycin, vancomycin, and ampicillin-gentamicin for treatment of experimental endocarditis caused by penicillin-resistant enterococci. Antimicrob Agents Chemother
39. Neu HC. Synergy and antagonism of combinations with quinolones. Eur J Clin Microbiol Infect Dis
40. Hu Y, Li L, Li W, et al. Combination antibiotic therapy versus monotherapy for Pseudomonas aeruginosa
bacteraemia: a meta-analysis of retrospective and prospective studies. Int J Antimicrob Agents
41. Vardakas KZ, Tansarli GS, Bliziotis IA, Falagas ME. beta-Lactam plus aminoglycoside or fluoroquinolone combination versus beta-lactam monotherapy for Pseudomonas aeruginosa
infections: a meta-analysis. Int J Antimicrob Agents
42. Poulikakos P, Tansarli GS, Falagas ME. Combination antibiotic treatment versus monotherapy for multidrug-resistant, extensively drug-resistant, and pandrug-resistant Acinetobacter
infections: a systematic review. Eur J Clin Microbiol Infect Dis
43. Bliziotis IA, Samonis G, Vardakas KZ, et al. Effect of aminoglycoside and beta-lactam combination therapy versus beta-lactam monotherapy on the emergence of antimicrobial resistance: a meta-analysis of randomized, controlled trials. Clin Infect Dis
44. Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the Infectious Diseases Society of America for the treatment of methicillin-resistant Staphylococcus aureus
infections in adults and children: executive summary. Clin Infect Dis
45. Brown KA, Khanafer N, Daneman N, Fisman DN. Meta-analysis of antibiotics and the risk of community-associated Clostridium difficile
infection. Antimicrob Agents Chemother
46. Martin-Loeches I, Perner A. Focus on infection and sepsis in intensive care patients. Intensive Care Med
47. Martin-Loeches I, Valles X, Menendez R, et al. Predicting treatment failure in patients with community acquired pneumonia: a case–control study. Respir Res
48. Martin-Loeches I, Deja M, Koulenti D, et al. Potentially resistant microorganisms in intubated patients with hospital-acquired pneumonia: the interaction of ecology, shock and risk factors. Intensive Care Med
2013; 39:672–681. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23358539
. [Cited 4 Mar 2013; Internet].
49. Sawyer RG, Claridge JA, Nathens AB, et al. Trial of short-course antimicrobial therapy for intraabdominal infection. N Engl J Med
50. Martin-Loeches I, Bos LD, Povoa P, et al. Tumor necrosis factor receptor 1 (TNFRI) for ventilator-associated pneumonia diagnosis by cytokine multiplex analysis. Intensive Care Med Exp
2015; 3:26Available from: http://www.ncbi.nlm.nih.gov/pubmed/26377207