EMPIRICAL ANTIBIOTIC THERAPY OF SEPSIS IN SURGICAL INTENSIVE CARE UNIT

Retrospective study was conducted in surgical intensive care unit (ICU) in Clinical Hospital Center Zagreb in . Th e aim of study was to create guidelines for empirical antibiotic therapy of sepsis in ICU for unknown causative agent based on antimicrobial susceptibility of causative bacteria. Th irty-two patients with severe sepsis were included in study and from medical records their clinical and microbiological data were analyzed. Antimicrobial susceptibility of the strains isolated from the bloodculture was tested by disk diff usion method according to CLSI (Clinical Laboratory Standard Institution). We used APACHE II score to predict the severity of illness. Mann-Whitney test and χ test were used to test statistical signifi cance diff erence between results. Acinetobacter baumannii and Pseudomonas aeruginosa were the predominant causative agent. Acinetobacter baumannii was displaying excellent susceptibility to ampicillin+sulbactam and carbapenems, whereas Pseudomonas aeruginosa was showed good susceptibility on ceftazidim and carbapenems. Methicillin-resistant Staphylococcus aureus (MRSA), third predominant causative agent exhibiting good susceptibility to vancomycin and linezolide. Th e recommended therapy is empirical antibiotic therapy and should cover all important pathogens.


Introduction
Th e International Sepsis Forum was published in  guidelines for the management of patients with severe sepsis and septic shock, which were including an evidence-based review on antibiotic therapy ().These guidelines have been updated and extended in , under the auspices of the Surviving Sepsis Campaign, critical care and infectious disease experts representing  international organizations ().Criteria for sepsis definition and antibiotic therapy pattern are recommended until now.Sever sepsis (infection-induced organ dysfunction or hypo-perfusion abnormalities), septic shock (hypotension not reversed with fluid resuscitation and associated with organ dysfunction or hypo-perfusion abnormalities) and mortality of sepsis remains unacceptably high in most ICU (, ).Infection in sepsis can occur at any site, most commonly in the respiratory tract, abdomen and bloodstream, and in more than  were caused by bacteria (, ).Thus, it is very important to find a precise microbiological diagnosis in patients with sepsis and to ensure effective antimicrobial therapy ().Th e appropriate empirical antibiotic therapy depends on following factors: patient's history (including drug intolerance), underlying diseases and susceptibility patterns of microorganisms in the hospital environment ().Choice of appropriate empirical antibiotic therapy in sepsis is one of the most important factors for the better outcome of patients in sepsis ().Th e initial empirical antimicrobial regimen should be broad enough to cover likely pathogens; for mixed (polymicrobial) or one causative agent infection.Mixed sepsis has been a more serious therapeutic problem.The aim of this study was to define and recommend the appropriate empirical antibiotic therapy for sepsis in surgical intensive care unit (ICU) in Clinical Hospital Center Zagreb.

Study population
 patients with positive blood-culture (BC) in surgical ICU in Clinical Hospital Center Zagreb in  were identifi ed from clinical microbiology laboratory records.Epidemiological, clinical and microbiological data was analyzed from the medical records of these  patients.Identification and antimicrobial susceptibility testing Blood cultures were processed with the BACTEC system and organisms were identi-fied using Vitek system.Results were interpreted according to the guidelines of the Clinical Laboratory Standard Institution (CLSI).Intermediate susceptibility to the antibiotics was considered as resistance.

Study design and data collection
The medical records of patients were retrospectively reviewed.The data collected included: age, gender, severity of illness (as calculated by acute physiology and chronic health evaluation-APACHE II score at a time of bacteriemia), antimicrobial therapy regimen at admission in ICU, changing of the therapy according to disk diffusion test (antibiogram), in vitro effectiveness of empirical antimicrobial agents, in vitro effectiveness of definitive antimicrobial agents and antimicrobial susceptibility of causative agents, type of infection (caused by one agent or polymicrobial), type of agents causing infection and outcome of patients.

Antibiotic therapy
Blood culture had been taken first, and than empirical antimicrobial therapy was initiated.Initiated therapy was changed to definitive therapy according to the results of culture and susceptibility testing within  days after the bacteriemia episode.The attending physician was decided about definitive therapy, mono-therapy vs. combined therapy.

Defi nitions
Positive blood-culture (BC) was defi ned as the isolation of the bacterial species from one or more blood cultures, and by the presence of clinical features consistent with sepsis.The standard definition for sepsis (severe sepsis and septic shock) provided by the Internal Sepsis Forum were used to determine septic patients.Severe sepsis was defi ned as sepsis associated with evidence of at last one organ dysfunction and positive bloodculture.Mixed sepsis was defined as the presence of two or more causative bacteria in the blood-culture.

Statistical analysis
Statistical significance difference between results was tested by Mann-Whitney test and χ  test.

Results
In a study  patients with severe sepsis were included.The mean age for all patients was  (range -),  male and  female.Sepsis caused by one agent were present at  () patients, and sepsis caused by more than one agent, polymicrobial (mixed) sepsis, were present at  () patients (Chart ).Nine different causative agents were isolated,  in mono-agent sepsis, and  in mixed sepsis (Chart , , ).
Acinetobacter baumannii was the predominant causative agent in both type of sepsis, it was found in  patients with mono-agent sepsis and in  patients showing mixed sepsis, whereas Pseudomonas aeruginosa was also presented as predominant causative agent in mono-agent sepsis, it was found at  patients, but only at  patients having mixed sepsis.Methicilin-resistant Staphylococcus aureus (MRSA) was identified at  patients showing monoagent sepsis and at  patients with mixed sepsis.Escherichia coli, Serratia marcescens and Enterococcus faecalis were detected with equal frequency in one () patient with mono-agent sepsis, respectively (Chart ).Escherichia coli were found in  () patients with mixed sepsis (Chart ).Klebsiella pneumoniae, Enterobacter spp., Staphylococcus epidermidis were detected with equal frequency in  () patients with mixed sepsis.Enterococcus faecalis was found at  () patient with mixed sepsis.The initial (empirical) antibiotic therapy was changed in  () patients, four of them died.Initial antibiotic therapy was not changed in  () patients, six of them died.Average value of APACHE II score in patients who died was  and was higher than in patients who were survived their APACHE II score was .Total mortality of  correlated with the predicted value of APACHE II score for mortality, which is  for APACHE II score of .

Antimicrobial susceptibility
All Acinetobacter baumannii strains were susceptible to ampicillin+sulbactam, imipenem, meropenem.Among Pseudomonas aeruginosa no resistance to ceftazidim, meropenem, colistin was observed. of methicillin-resistant Staphylococcus aureus (MRSA) strains were susceptable to vancomicin and rifampicin.All Enterobacteriaceae (Klebsiella pneumoniae, Escherichia coli and Enterobacter spp.) strains were sensitive to ceftazidim, gentamicin, imipenem and meropenem.Staphylococcus epidermidis was sensitive to ampicillin and cloxacillin.Enterococcus spp. was sensitive to ampicillin, vancomycin and rifampicin.

Statistical analysis
Th ere was statistically signifi cant diff erence in APACHE II score between patients who were survived and those who died (p<,).No signifi cant diff erence was found in APPACHE II score between patients who died with mono-agent sepsis and mixed sepsis (p>,).According to antibiogram antimicrobial therapy was significantly more often changed for younger patients compared to the older patients (p<,).Correction of the therapy, related to a causative agent, was more often necessary for mixed sepsis than mono sepsis (p<,).

Discussion
Sepsis is still associated with significant mortality and high health-care costs in spite of the all known guidelines and recommendation for treating ().The many recent studies report high frequency mortality rates of sepsis, from  of patients dying in the ICU, rising to > in patients with severe sepsis and septic shock (,).Th ere are also recent studies reports where mortality rate is not so high, for example, the EPISEPSIS study in France reported that  of patients had severe sepsis and UK study reported a rate of  ().Th is study included patients admitted for routine postoperative surveillance who are likely to have lower rates of complications.Many of our patients were admitted in emergency, so that higher mortality rate of  is in concordance with the present bibliographic data.
Furthermore, this proportion remains high in spite of the new antimicrobial agents (, ).Type of sepsis is another important problem: mono-agent sepsis (sepsis caused with one agent) presents less therapeutic problem than mixed sepsis (sepsis caused with two or more agent).Th ere was  mixed sepsis and  mono-agent sepsis in our study.Th is is also correlated with the data published by other authors (, , ).Th ere was no signifi cant diff erence between patients who died due to the mono-agent sepsis and mixed sepsis (p >,).APACHE II score remains reliable predictive data for patient's outcome (, , , ).Therapeutic approach for intraabdominal infection, pneumonia (ventilator associated pneumonia -VAP) or catheter-associated infection is different, so identification of sepsis's source (focus) could be very helpful, especially for the empirical therapy ().Intraabdominal infection, followed by VAP was the predominant source of sepsis in our surgical ICU.Some authors suggested fl ucloxacillin + amikacin like empirical therapy for unknown causative agent ().Paul-Erlich-Gesellschaft recommended cephaslosporine third generation and aminoglicoside (, ).According to the antibiograms of the strains isolated in our surgical intensive care unit ICU, we would suggest cephalosporine third generation and ampicillin+sulbactam for the empirical therapy of the sepsis with the unknown causative agent.Since staphylococci were isolated from blood-culture in some of our patients, in our opinion beta-lactamase stable penicillin (fl ucloxacillin) should be added to empirical therapy ().Acinetobacter baummanii and Pseudomonas aeruginosa often require administration of carbapenems, due to their multiple antibiotic resistances (, ).Carbapenems remains antibiotic of choice for the treatment of infections caused by Klebsiella pneumoniae and Escherichia coli.Sepsis caused with methicillin-resistant Staphylococcus aureus (MRSA) should be treated with vancomycin or linezolide (, ).