Klebsiella pneumoniae Infections in the Intensive Care Unit: Risk Factors Related to Carbapenem Resistance and Patient Mortality

BÜYÜKTUNA S. A. , HASBEK M., ÇELİK C., Unlusavuran M., AVCI O., BALTACI S., ...More

MIKROBIYOLOJI BULTENI, vol.54, no.3, pp.378-391, 2020 (Journal Indexed in SCI) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 54 Issue: 3
  • Publication Date: 2020
  • Doi Number: 10.5578/mb.69679
  • Title of Journal : MIKROBIYOLOJI BULTENI
  • Page Numbers: pp.378-391


Klebsiella pneumoniae is the cause of complicated and difficult-to-treat nosocomial infections such as sepsis, urinary tract infection, catheter related infections, pneumonia and surgical site infections in intensive care units. The biggest problem in infections with K.pneumoniae is that treatment options are limited due to multiple antibiotic resistance and consequently the increased morbidity and mortality. The widespread and improper use of carbapenems can lead to epidemics that are difficult to control, especially in intensive care units because of the acquired resistance to this group of antibiotics. Outbreaks and sporadic cases caused by carbapenem resistant K.pneumoniae (CRKP) species have been reported all over the world in recent years with increased frequency. The aim of this study was to determine the risk factors related to carbepenem resistance and mortality caused by K.pneumoniae infections in a university hospital anesthesia intensive care unit. The study was conducted between January 1st, 2016, and December 31st, 2018. Retrospective data were obtained from the patient and laboratory-based surveillance records. Adult patients (>= 18 years) with K.pneumoniae growth in the blood, urine, abscess and tracheal aspirate samples collected 48 hours after admission to the intensive care unit were considered as the relevant infection locus-related agent and treated with antibacterial therapy. Clinical samples collected from patients were inoculated onto 5% sheep blood and eosin-methylene-blue (EMB) agar except the blood samples. Blood samples were cultured in blood culture bottles and incubated in an automated system. Gram staining was performed for the samples showing growth signal within five days and then inoculated onto 5% sheep blood and EMB agar media and were incubated for 18-24 hours at 35.5-37 degrees C. Identification of the isolates was performed using Bruker IVD MALDI Biotyper 2.3 (Bruker Daltonik GmbH, Bremen, Almanya) based on "matrix-assisted laser desorption/ionization time-of-mass spectrometry (MALDI-TOF MS)". K.pneumoniae isolates were identified by obtaining reliability scores of 2.0 and above in the study. Antibiotic susceptibility tests were performed with Phoenix 100 (BD, New Jersey, ABD) automated system. Interpretations were made according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines. Combination disk diffusion test and polymerase chain reaction based tests were used to show the presence of carbapenemase in CRKP isolates. A total of 88 patients with K.pneumoniae infection were included in the study. The mean age of the patients was 74 +/- 15 (range= 21-93) years and 60.2% were female. CRKP was detected in 32 patients (36.4%) and carbapenem-sensitive K.pneumoniae (CSKP) was detected in 56 patients. The presence of OXA-48 was found to be 68.8% in the carbapenem screening test performed by combination disc method in patients with CRKP. Multivariate logistic regression analysis showed that previous use of colistin [Odds ratio (OR)= 19.108; 95% confidence interval (CI). 2.027-180.133; p= 0.010] and aminoglycoside (OR= 12.189; 95% CI= 1.256-118.334; p= 0.031) was an independent risk factor in terms of CRCP among the patients with K.pneumoniae infection. The 28-day mortality rates were 71.9% in the CRKP group (23/32) and 37.5% in the CSKP group (21/56). Presence of CRKP in terms of 28-day mortality (OR= 5.146; 95% CI= 1.839-14.398; p= 0.002) was an independent risk factor.