Summary
Introduction: Bacteria of the genus Klebsiella are important causes of nosocomial infections. These bacteria may develop resistance in different ways, with extended-spectrum beta-lactamase (ESBL) production being the most common. The aim of this study was to evaluate resistance status of Klebsiella isolates obtained from different intensive care units and various clinical samples.
Materials and Methods: Resistance rates of Klebsiella spp. strains isolated from bacteriologic cultures of patients in intensive care units between 2014 and the first six months of 2017 were evaluated retrospectively. Blood, urine, bronchoalveolar lavage, cerebrospinal fluid, and tracheal aspirate samples were cultured. Intensive care patients aged ≥18 years were included. Only the first isolate from each patient was included; subsequent isolates from the same patient were excluded.
Results:A total of 443 patients were included in the study. Of the Klebsiella-positive samples, 31.8% (n=141) were tracheal aspirate, 23.7% (n=105) were blood, 23.3% (n=103) were urine, 9.0% (n=40) were sputum, and 12.2% (n=54) were other samples. The majority of cases were K. pneumoniae (89.7%, n=397), followed by K. oxytoca (7%, n=31), K. ozaenae (2.9%, n=13), K. granulomatis (0.2%, n=1), and K. ornithinolytica (0.2%, n=1). Of the 443 isolates, 54.4% (n=241) were ESBL-producing while 45.6% (n=202) were non-ESBL-producing. Mortality rates were 60% among patients with ESBL-positive Klebsiella and 47.5% among patients with ESBL-negative Klebsiella (p=0.008).
Conclusion: The antibiotic resistance of ESBL-positive Klebsiella isolates in our hospital was higher than resistance rates in the literature. To overcome this resistance issue, each hospital must know its own resistance rates and establish policies for the rational use of antibiotics.
Introduction
Klebsiella bacteria may develop different mechanisms of antibiotic resistance. Among them, the most common is extended-spectrum beta-lactamase (ESBL) production[5]. Another form is carbapenem resistance[6]. Resistant bacteria lead to prolonged hospitalization, loss of labor, high costs, and high mortality.
Intensive care patients are critical patterns who require early and effective treatment. Therefore, each hospital must be aware of its specific bacterial resistance rates and initiate empirical treatment accordingly. The aim of this study was to determine the resistance profiles of Klebsiella bacteria isolated from various sample types in various clinical samples in the intensive care units (ICUs) of our setting and contribute to policies regarding the appropriate use of antibiotics.
Methods
Clinical samples received by the microbiology laboratory were inoculated on 5% sheep blood agar (RTA, Turkey) and EMB agar (RTA, Turkey) and incubated at 37 oC for 18-24 hours. Colonies showing Gram-negative bacilli in Gram staining and were identified as oxidase negative, lactose positive, and mucoid producing on culture medium were suspended at a turbidity of 0.5 McFarland. Identification and antibiotic susceptibility tests were done in accordance with "European Committee on Antimicrobial Susceptibility Testing" (EUCAST) recommendations using a Phoenix 100 (Becton Dickinson, USA) system. Extended-spectrum beta-lactamase-producing organisms were identified based on susceptibility to aztreonam, third and fourth generation cephalosporins (cefpodoxime 8 μg/mL, ceftazidime 8 μg/mL) alone and/or in the presence of beta-lactamase inhibitors (ceftriaxone/clavulanic acid 2 μg/ mL, cefotaxime/clavulanic acid 2 μg/mL, ceftazidime/clavulanic acid 2 μg/mL). Imipenem, meropenem, and ertapenem minimum inhibitor concentration values were tested and metallo-betalactamase producing strains were considered carbapenemresistant.
Statistical Analysis
SPSS v.17.0 software package (SPSS Inc, Chicago, Illinois, USA) was used in statistical analyses of the data obtained from the study. Continuous data were summarized as mean and standard deviation, while categorical data were expressed as number and percentage. Student"s t-test was used in comparisons of continuous variables between independent groups. The chisquare (χ²) test was used when comparing categorical values between independent groups. Statistical significance level was accepted as p<0.05.
Results
Of the samples from which Klebsiella was isolated, 31.8% (n=141) were TAC, 23.7% (n=105) were blood, 23.3% (n=103) were urine, 9.0% (n=40) were sputum, and 12.2% (n=54) were other sample types.
The three most common units where Klebsiella-positive cultures were performed were the Anesthesia ICU with 56.9% (n=252), the Internal Medicine ICU with 16.9% (n=75), and the Neurology ICU with 10.4% (n=46). Data pertaining to all units are summarized in Table 1.
In terms of the distribution of Klebsiella species, Klebsiella pneumoniae accounted for the large majority of cases (89.7%, n=397). This was followed by K. oxytoca (7%, n=31), K. ozaenae (2.9%, n=13), K. granulomatis (0.2%, n=1), and K. ornithinolytica (0.2%, n=1).
Extended-spectrum beta-lactamase-positive strains accounted for 54.4% (n=241) of the cases, while 45.6% (n=202) were ESBL-negative.
The antibiotic susceptibility rates of the K. pneumoniae and K. oxytoca strains are presented in Table 2.
The mean duration of hospital stay was 47.04±40.7 days (minimum-maximum, 1-220 days). Hospital stays were slightly longer in ESBL-positive cases (47.65±39.6 days) compared to ESBL-negative cases (46.31±42.0 days), but the difference was not statistically significant (p=0.729).
The mortality rate was 54.4% while the remaining 45.6% of the patients were discharged with full recovery. All deaths occurred at the hospital, on day 42.3±34.7 of hospitalization. The mortality rate was 60.2% among patients with ESBL-positive Klebsiella versus 47.5% among patients with ESBL-negative Klebsiella (χ², p=0.008). In addition, the mortality rate was 55.7% among patients with carbapenem-resistant Klebsiella and 53.8% among patients with non-carbapenem-resistant Klebsiella (χ², p=0.717). The association between culture types and patient mortality is shown in Table 3.
Discussion
While the prevalence of ESBL-producing Klebsiella is increasing in Turkey and eastern Europe, it is on a downward trend in western Europe. Antimicrobial resistance patterns may vary between regions, hospitals, and even units within the same hospital. Therefore, each hospital must know its specific antimicrobial resistance profile. Rates of ESBL-producing Klebsiella vary widely in the literature, too. Temiz et al.[12] detected ESBL-producing Klebsiella at rates of 65.1% in samples obtained from ICUs, inpatient units, and outpatient units (68.3% in K. pneumoniae isolates and 10.7% in K. oxytoca isolates). Parlak et al.[13] reported ESBL-positive K. pneumoniae in 67% of samples obtained from inpatient units and the ICU. Güdücüoğlu et al.[14] reported this rate to be 49%. In the HITIT-2 study involving 6 centers in Turkey, the prevalence of ESBL-producing K. pneumoniae was 32.3%, with only 16.2% in ICUs[15]. The International Nosocomial Infection Control Consortium report for Turkey summarized data from 19 hospitals between 2003 and 2012[16]. It was determined in the study that ceftriaxoneand ceftazidime-resistant K. pneumoniae strains accounted for 55.7% of central catheter-related bloodstream infections, 46.3% of ventilator-associated pneumonia, and 50.0% of catheter-related urinary tract infections in the ICU. The ESBLpositive Klebsiella ratio in our study was 54.4%.
Beta-lactam/beta-lactamase inhibitors may be preferred for infections caused by ESBL-producing bacteria. In a study by Kuzucu et al.[17], 68% of ESBL-producing K. pneumoniae isolates were found to be piperacillin/tazobactam resistant. This rate was 32% in another study[18], and was 22.3% in the HITIT-2 study[15]. The results of our study were similar to those of Kuzucu et al.[17]. Resistance to beta-lactam/beta-lactamase inhibitors is increasing.
In severe/systemic infections caused by ESBL-producing Klebsiella spp., drugs should be chosen based on susceptibility patterns. In their 2005-2006 analysis, Güdücüoğlu et al.[14] observed no carbapenem resistance among ESBL-positive K. pneumoniae isolates. There was also no resistance to carbapenem in three of the centers (Akdeniz, Hacettepe, and Dokuz Eylül Universities) involved in the multicenter HITIT-2 study conducted in 2004- 2005. However, carbapenem resistance was detected at a rate of 1.3% in the other three centers[15]. More recently, a study by Saygılı Pekintürk and Akgüneş[19] reported imipenem resistance rates of 2%, 7%, 0%, 38%, and 50% respectively in the years 2011-2015. Although numerous previous studies demonstrated carbapenem susceptibility of ESBL-positive Klebsiella isolates, current evidence shows that carbapenem resistance is becoming more prevalent among ESBL-positive Klebsiella isolates. We also found high resistance to carbapenems in the ESBL-positive K. pneumoniae strains in our study, with rates of 40.3% for ertapenem, 30.5% for meropenem, and 29.7% for imipenem.
In ESBL-positive infections, drugs in the quinolone group may be used as an alternative to beta-lactam/beta-lactamase inhibitors or carbapenems[20]. However, previous studies have generally reported high quinolone resistance rates. Parlak et al.[13] determined ciprofloxacin resistance rates of 11%, 8%, 13%, 35%, and 36% in ESBL-positive K. pneumoniae isolates between 2006 and 2010. In another study, the average ciprofloxacin resistance in the years 2011-2015 was 46% (no data for 2011; 48%, 24%, 52%, and 58% respectively for the remaining years) [19]. In another study by Nepal et al.[21], ciprofloxacin resistance of ESBL-positive K. pneumoniae isolates was 46.2%. In our study, the rate of ciprofloxacin resistance in ESBL-positive K. pneumoniae isolates was much higher than the rates reported in the literature.
We detected higher rates of antibiotic resistance in the ESBLpositive Klebsiella isolates in our study compared to resistance rates in the literature. This may be attributed to the fact that all of the patients included in this study were ICU patients.
Conclusion
Ethics
Ethics Committee Approval: Retrospective study.
Informed Consent: Retrospective study.
Peer-review: Externally and internally peer-reviewed.
Authorship Contributions
Concept: S.N., S.A., Design: S.N., S.A., A.R.Ş., Data Collection or Processing: B.T., S.N., Analysis or Interpretation: S.N., A.R.Ş.,
Literature Search: S.N., A.R.Ş., Writing: S.N., S.A.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.
