Ventilator-associated pneumonia (VAP) occurs in 10-20% of patients mechanically ventilated for more than 48 h and is associated with the mortality rate of 20-50%, or up to 70% in Pseudomonas aeruginosa and Acinetobacter baumannii infections [1].

High percentages of therapy failures in such patients are associated with improper choice of antibiotics for empiric therapy, whose spectrum does not cover the wide range of microorganisms and their mechanisms of resistance. Empiric antibiotic therapy should follow the generally accepted current recommendations and consider the information about local pathogens and their susceptibility. The effectiveness of individual therapy regimens differs according to the country, region or health institution, and depends markedly on the spectrum of microorganisms, antibiotics used earlier and the resultant mechanisms of resistance.

The aim of the present study was to determine the spectrum of susceptibility of Klebsiella, Pseudomonas and Acinetobacter strains, isolated from VAP patients in relation to the recommended regimens of empiric antibiotic therapy.

METHODS

Samples from bronchoalveolar lavages (BALs) were collected from mechanically ventilated patients with VAP, diagnosed according to the clinical and microbiological criteria, between February and September 2010. The patients were hospitalized in the Department of Intensive Therapy, surgical departments of various profiles and in the Department of Lung Diseases of the Teaching Hospital no.4 in Lublin. In total, 81 Gram-negative Klebsiella, Pseudomonas and Acinetobacter strains were analysed.

The material was routinely cultured on the Columbia agar with 5% of sheep blood, mannitol salt agar, MacConkey agar and Sabouraud agar. They were incubated at 35o C for 24-72 h under oxygen conditions and  the minimal inhibitory concentrations (MICs) for the antibiotics recommended for empiric therapy in VAP were determined. MICs for imipenem, meropenem, cefepime, ceftazidime, piperacillin with tazobactam, ciprofloxacin, amikacin and gentamicin were determined using an automated VITEK 2 system; in the case of doripenem, the Etest assay with AB BIODISK strips on the Mueller Hinton agar was applied. Incubation was carried out at 35˚C for 16-20 h under oxygen conditions. The MIC values were interpreted according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) and defined as susceptible (S), intermediate (I) and resistant (R).

RESULTS

The following bacterial strains were identified in the collected material: A. baumannii (34), Klebsiella (25, including 24 Klebsiella pneumoniae and 1 Klebsiella oxytoca) and Pseudomonas aeruginosa (22). Klebsiella (17) and A. baumannii (14) strains were predominant at the Department of Intensive Therapy, the number of P. aeruginosa spp. was lower (7); in the surgical departments A. baumannii (14) and P. aeruginosa (12) were mainly detected followed by Klebsiella strains (6), whereas in the Department of Lung Diseases, more A. baumannii (6) than P. aeruginosa (3) and Klebsiella (2) strains were identified.

The individual strains showed varied susceptibility. Klebsiella spp. showed 100% susceptibility to all carbapenems. A high susceptibility was also found amongst P. aeruginosa strains; 81.8%, 73.7% and 68.4% of susceptible strains in vitro for doripenem, meropenem and imipenem, respectively. The susceptibility of A. baumannii strains to doripenem was only 29.4%, at relatively high susceptibility to meropenem and imipenem, 71% and 72.4% (Table 1).

The highest percentages of strains susceptible to carbapenems in monotherapy were 64.6% for doripenem, 80.6% for meropenem and 80% for imipenem (Fig. 1).

P. aeruginosa strains were found to be susceptible to cefepime (71.4%); amongst Klebsiella and A. baumannii spp. only 58.3% and 40% of strains were susceptible to this antibiotic, respectively. A high percentage of P. aeruginosa strains was susceptible to ceftazidime (81.8%) compared to lower percentages of A. baumannii and Klebsiella spp. – 5.9% and 32%, respectively (Table 2). The cephalosporins recommended for monotherapy were characterized by lower percentages of strains susceptible to carbapenems: 60.5% for cefepime and 34.6% for ceftazidime in the entire pool of microorganisms (Fig. 2).

The lowest in vitro activity was observed in combined regimens based on β-lactam antibiotics with inhibitors, with higher percentages of susceptible strains in the β-lactam/inhibitor and aminoglycoside regimen compared to β-lactam/inhibitor and fluoroquinolone therapy. Amongst A. baumanni, an extremely low percentage of strains susceptible to both drugs was found: 9.1%, 3.3% and 0% for the combination of β-lactam/inhibitor with amikacin, gentamicin and ciprofloxacin, respectively; a number of strains were resistant to both antibiotics. The same regimens gave better results for Klebsiella spp.: for the combination of β-lactam/inhibitor with amikacin, gentamicin and ciprofloxacin, 58.3%, 34.8% and 28.6% of strains susceptible to both antibiotics were detected. The best results in vitro were observed for P. aeruginosa: for the combination of β-lactam/inhibitor with amikacin, gentamicin and ciprofloxacin, 81%, 63.6% and 72.7% of strains susceptible to both antibiotics were demonstrated (Table 3). The total percen
tage of strains susceptible to both antibiotics in piperacillin/tazobactam and amikacin regimens was 56.8% and with gentamicin 30.7%; in the piperacillin/tazobactam and ciprofloxacin regimens, this percentage was only 29.7% (Fig. 3).

DISCUSSION

Antibiotic therapy in VAP is a compromise between appropriate and early pharmacotherapy based on broad-spectrum antibiotics, knowledge of local epidemiology and susceptibility of bacteria and fear of increasing resistance to the antibiotics presently available. Both delayed treatment and improper therapy (an antibiotic ineffective against a VAP-inducing pathogen, improper dose or pharmacodynamic parameters, duration of therapy) are associated with worse outcomes – higher incidence of septic shock and higher mortality [2, 3].

According to the current national and international recommendations, mono- or multidrug regimens should be used in VAP of late onset or with risk factors of resistant pathogens. In its guidelines for diagnosis and early treatment of respiratory infections, the Polish Society of Lung Diseases recommends the empiric broad-spectrum antibiotic therapy - cephalosporins active against Pseudomonas (cefepime, ceftazidime) or carbapenems active against Pseudomonas (imipenem, meropenem), or combined therapy – β-lactam with inhibitors (piperacillin/tazobactam, titarcillin/clavulanate) in combination with fluoroquinolones active against Pseudomonas (ciprofloxacin) or aminoglycosides (amikacin, gentamicin) [4].

In our study, the highest in vitro activity amongst carbapenems was found for meropenem and imipenem. The microorganism susceptibility was comparable when interpreted according to CLSI. Doripenem showed lower percentages of susceptible strains in vitro, which should be related to the presence of A. baumannii resistant to this antibiotic (70.6% within the species) and preserved susceptibility to the remaining carbapenems. In the pool of microorganisms examined, the number of A. baumannii, P. aeruginosa and Klebsiella spp. was comparable. Under such circumstances (lacking the predominating pathogen and considering empiric therapy), the susceptibility of the entire pool was analysed. A high percentage P. aeruginosa strains susceptible to all carbapenems, especially doripenem, is worth noticing. It should be remembered, however, that this microorganism may become unsusceptible during carbapenems therapy in vivo, e.g. due to loss of the protein D2 involved in transport of carbapenems or in active removal of these drugs from the cell [5].

Carbapenems are the commonest antibiotics used in VAP in Europe; their use increases when the VAP inducing pathogens contain Acinetobacter spp. By more than 10% [6]. The clinical efficacy of meropenem was demonstrated in the prospective, randomized study involving 140 patients with the diagnosis of VAP. In the group receiving meropenem positive responses to treatment were observed in 82.5% of cases [7]. The studies regarding imipenem and doripenem revealed that their clinical efficacy was 64.2% and 68.3%, respectively [8].

In the hospital wards, from which the strains examined by us were taken, empiric therapy of VAP was based on carbapenems. The findings confirm their superiority in empiric VAP therapy. Moreover, doripenem, a novel drug from the carbapenems group, may be found useful in the treatment of such infections. Although our results showed its lower activity in vitro against the examined strains compared to the remaining carbapenems, doripenem was found to be more active when compared with the other regimens of mono- and multi-drug therapy. The activity of doripenem against P. aeruginosa and Klebsiella spp. was comparable yet much lower activity against A. baumannii spp. was observed. Its better pharmacodynamic and pharmacokinetic properties, stability in infusion solutions up to 12 h, possible administration in the long-lasting infusion, which enables the maintenance of concentration >MIC at the infection site and lower capacity to induce resistance against the remaining carbapenems and lower neurotoxicity, are in fav
our of doripenem [9, 10, 11].

According to the recommendations of the Polish Society of Lung Diseases, an alternative to monotherapy with carbapenems is the use cephalosporins active against Pseudomonas. Our findings do not confirm the effectiveness of such a choice – high percentage of strains resistant in vitro indicated the risk of treatment failures in the clinical setting due to a high percentage of strains resistant to microorganisms belonging to other species. The in vitro activity of cephalosporins against Pseudomonas was higher than against the remaining pathogens and comparable with the high activity of carbapenems. The recommendations of the American Thoracic Society/Infectious Diseases Society of America and European Society of Intensive Care Medicine/European Society of Clinical Microbiology and Infectious Diseases/European Respiratory Society suggest the combination of cephalosporins with aminoglycosides or fluoroquinolones [12, 13]; however, many studies demonstrate limited success of this management [14, 15]. In the clinical practice, combined therapy is not associated with higher percentages of success but with the increased risk of selection of resistant strains and superinfections, hence higher risk of adverse toxic effects of the antibiotics used.

Among the Klebsiella strains examined, 68% produced extended-spectrum ß-lactamases (ESBL) whereas strains producing carbapenems were not detected. According to the Beta-P1 Study Group, the percentage of ESBL-producing strains amongst Enterobacteriacae in Polish hospitals is 11.1% [16]. In clinical practice, the occurrence of ESBL provides information about therapeutic limitations and likely clinical effectiveness of cephalosporins. The clinical effectiveness is affected by numerous factors depending on the pathogen’s features, clinical condition of a patient and an antibiotic chosen. When ESBL strains are present, the drugs of choice are carbapenems; piperacillin with tazobactam, aminoglycosides and fluoroquinolones can also be effective [17]. Our findings confirm the susceptibility of the examined stains to all carbapenems (100% of susceptible strains). The use of piperacillin with tazobactam, aminoglycosides and fluoroquinolones in the treatment of ESBL strains should be limited to the targeted therapy with
monitoring of MIC, pharmacokinetic and inflammatory parameters, as the markers of clinical efficacy.

The in vitro activity of cephalosporins against the isolated strains was lower than that of carbapenems. High percentages of strains resistant to ceftazidime are connected with the fact that 68% of Klebsiella strains produce ESBL. It is increasingly stressed that clinical efficacy of third- and forth-generation cephalosporins depends on MIC rather than the enzyme production, therefore strains of MIC below the border point may be considered susceptible, irrespectively of ESBL production [18]. Our experiences confirm that such management is grounded. Almost 1/3 of ECBL-producing Klebsiella strains were also susceptible to cefepime, which would not have been used if the therapy had been based on the criterion of enzyme production instead of MIC. A high proportion of unsusceptible Acinetobacter markedly affected a high percentage of strains resistant to ceftazidime.

The recommendations regarding the initial empiric antibiotic therapy have to be modified according to the knowledge about pathogens predominating in a given centre and local data on antibiotic resistance. In order to undertake proper empiric treatment, hospital wards should closely cooperate with the microbiological laboratory and design their own empiric formularies for a particular ward, irrespective of the available literature recommendations. Such empiric formularies based on BAL microbiological studies resulted in better  outcomes of VAP therapy (85%) than those based on current recommendations (73%) [19].

CONCLUSIONS

1. Monotherapy in VAP appears to be more effective than combined therapy.

2. In the pool of VAP pathogens examined, carbapenems (imipenem, meropenem, doripenem) seem to be the best drugs for empiric therapy.

3. Doripenem is active in vitro against P. aeruginosa and Klebsiella spp. and can be included in the initial set of antibiotics for empiric VAP therapy. In the environments of high incidence of A. baumannii strains, doripenem is probably not the drug of proper choice.

4. The in vitro activity of all examined antibiotics against P. aeruginosa is comparable, both in mono- and multi-drug regimens.

5. The diverse profile of pathogens isolated from VAP patients in the individual hospital wards and the wide range of susceptibility to empiric antibiotic therapy suggest the need to modify the current recommendations and to introduce new guidelines for hospitals/wards based on local microbiological studies, considering individual risk factors of patients.

..............................................................................................................................................................

REFERENCES

1.    Kollef MH, Shorr A, Tabak YP, Gupta V, Liu LZ, Johannes RZ: Epidemiology and outcomes of health–care associated pneumonia: results from a large US database of culture positive pneumonia. Chest 2005; 128: 3854-3862.

2.    Kollef MH, Micek ST: Standarization of care to improve outcomes of patients with ventilator-associated pneumonia and severe sepsis; in: Year Book of Intensive Care and Emergency Medicine (Ed.: Vincent JL), Springer-Verlag Berlin Heidelberg New York 2010; 235-243.

3.    Magnotti LJ, Schroeppel TJ, Fabian TC, Clement LP, Swanson JM, Fischer PE, Bee TK, Maish GO 3rd, Minard G, Zarzaur BL, Croce MA: Reduction in inadequate empiric antibiotic therapy for ventilator-associated pneumonia: impact of a unit-specific treatment pathway. Am Surg 2008; 74: 516-523.

4.    Komisja Zakażeń Układu Oddechowego Polskiego Towarzystwa Chorób Płuc. Zakażenia układu oddechowego: wytyczne rozpoznawania i leczenia. Polskie Towarzystwo Chorób Płuc, Katowice 2010; 37-38.

5.    Dzierżanowska  D: Patogeny bakteryjne zakażeń szpitalnych, w: Zakażenia szpitalne (Red.: Dzierżanowska D), α-medica press, Bielsko-Biała, 2008, 48-51.

6.    Rello J, Ulldemolins M, Lisboa T, Koulenti D, Mañez R, Martin-Loeches I, De Waele JJ, Putensen C, Guven M, Deja M, Diaz E, EU-VAP/CAP Study Group: Determinants of choice and prescription patterns in empiric antibiotic therapy for HAP/VAP. Eur Respir J 2010; 36: doi: 10.1183/09031936.00093010.

7.    Alvarez LF, Serious Infection Study Group: Efficacy of meropenem as monotherapy in the treatment of ventilator-associated pneumonia. J Chemother 2001; 13: 70-81.

8.    Chastre E: Efficacy and safety of intravenous infusion of doripenem versus imipenem in ventilator-associated pneumonia: a multicenter, randomized study. Crit Care Med 2008; 36: 1089-1096.

9.    Drzewiecki A: Doripenem w świetle badania COMPACT. Zakażenia 2009; 6: 38-41.

10.    Psathas PA, Kuzmission A, Ikeda K, Yasuo S: Stability of doripenem in vitro in representative infusion solutions and infusion bags. Clin Ther 2008; 30: 2075-2087.

11.    Tokarz A, Gaszyński T, Gaszyński W: Blaski i cienie antybiotykoterapii o szerokim spektrum – nowe karbapenemy. Zakażenia 2009; 5: 40-45.

12.    Guidelines for management of adults with hospital-acquired ventilator-associated and health care associated pneumonia. Am J Respir Crit Care Med 2005; 171: 388-416.

13.    Torres A, Ewig S, Lode H, Carlet J: For the European HAP working group: defining treating and preventing hospital acquired pneumonia: European perspective. Intensive Care Med 2009; 35: 9-29.

14.    Aarts MAW, Hancock JN, Heyland D, McLeod RS, Marshall JC: Empiric antibiotic therapy for suspected ventilator-associated pneumonia: a systematic review and meta-analysis of randomized trials. Crit Care Med 2008; 36: 108-117.

15.    Damas P, Garweg C, Monchi M, Nys M, Canivet JL, Ledoux D, Preiser JC: Combination therapy versus monotherapy: a randomized pilot study on the evolution of inflammatory parameters after ventilator associated pneumonia. Crit Care 2006; 10: R52.

16.    Empel J, Baraniak A, Literacka, Mrówka A, Fiett J, Sadowy E, Hryniewicz W, Gniadkowski M: Beta-P1 study Group: molecular survey of betalactamases conferring resistance to newer betalactams in Enterobacteriaceae isolates from Polish hospitals. Antimicrob Agents Chemother 2008; 52: 2449-2454.

17.    Duszyńska W: Antybiotykoterapia zakażeń bakteriami Gram-ujemnymi u chorych w OIT w kontekście wielooporności. Anaesthesiol Intensive Ther 2010; 3: 160-166.

18.    Ziółkowski G, Ziółkowska B: Wpływ antybiotykoterapii na wielooporność patogenów bakteryjnych. Sepsis 2010; 2: 155-164.

19.    Vincent JL, Barros DS, Cianferoni S: Diagnosis, management and prevention of ventilator-associated pneumonia: an update. Drugs 2010; 70: 1927-1944.

..............................................................................................................................................................

address:

*Maria Kozioł-Montewka
Zakład Mikrobiologii Lekarskiej
Uniwersytet Medyczny w Lublinie
e-mail: koziolm@yahoo.com

received:  14.03.2011
accepted:  15.07.2011