Klebsiella ESBL bacteremia-mortality and risk factors

Abstract

Background: Extended spectrum β-lactamase (ESBL)-producing bacteria have become recognized as a problem in South America. The aim of this study was to evaluate risk factors and mortality rate in bacteremia caused by ESBL-producing Klebsiella pneumoniae in a Brazilian hospital. Methods: A three-year retrospective cohort study with 104 cases of K. pneumoniae bacteremia (61 ESBL and 43 non-ESBL). Several clinical and laboratory variables were evaluated. The outcome of interest was 30-day mortality. The adequate treatment was evaluated according to antibiotic susceptibility. Results: Multivariable analysis showed that central venous catheter and mechanical ventilation were independent risk factors for ESBL. The duration of hospitalization before the bacteremia was not a risk factor. Mortality was similar in ESBL and non-ESBL and inadequate therapy was not shown to be a risk factor. Conclusion: ESBL-producing Klebsiella bacteremia can occur early, suggesting that a carbapenem should be included in the initial empirical therapy for bacteremia in patients under mechanical ventilation and/or central venous catheter in our institution.

Klebsiella pneumoniae; beta-lactamases; bacteremia; carbapenems


BRIEF COMMUNICATION

Klebsiella ESBL bacteremia-mortality and risk factors

Felipe Francisco TuonI; Margot KrugerII; Marcos TerreriII; Sergio R. Penteado-FilhoII; Lucas GortzIII

IMD, PhD, Professor, Division of Infectious and Parasitic Diseases, Hospital Universitário Evangélico de Curitiba, Curitiba, PR, Brazil

IIMD, Division of Infectious and Parasitic Diseases, Hospital Universitário Evangélico de Curitiba, Curitiba, PR, Brazil

IIIMSc, Division of Infectious and Parasitic Diseases, Hospital Universitário Evangélico de Curitiba, Curitiba, PR, Brazil

Correspondence to

ABSTRACT

Background: Extended spectrum β-lactamase (ESBL)-producing bacteria have become recognized as a problem in South America. The aim of this study was to evaluate risk factors and mortality rate in bacteremia caused by ESBL-producing Klebsiella pneumoniae in a Brazilian hospital. Methods: A three-year retrospective cohort study with 104 cases of K. pneumoniae bacteremia (61 ESBL and 43 non-ESBL). Several clinical and laboratory variables were evaluated. The outcome of interest was 30-day mortality. The adequate treatment was evaluated according to antibiotic susceptibility. Results: Multivariable analysis showed that central venous catheter and mechanical ventilation were independent risk factors for ESBL. The duration of hospitalization before the bacteremia was not a risk factor. Mortality was similar in ESBL and non-ESBL and inadequate therapy was not shown to be a risk factor. Conclusion: ESBL-producing Klebsiella bacteremia can occur early, suggesting that a carbapenem should be included in the initial empirical therapy for bacteremia in patients under mechanical ventilation and/or central venous catheter in our institution.

Keywords:Klebsiella pneumoniae; beta-lactamases; bacteremia; carbapenems.

INTRODUCTION

Gram-negative rods are the leading cause of bloodstream infection in Brazilian hospitals with increasing incidence of extended spectrum β-lactamase (ESBL)-producing bacteria.1Klebsiella pneumoniae is the most important ESBL-producing bacteria, and the incidence has reached worrisome levels and continues to increase in South America.2

Mortality rate is evidently associated with inadequate empirical antibiotic choice for ESBL bacteria. Carbapenems are the drugs of choice for treatment of ESBL-producing K. pneumoniniae (ESBL-KP), but this class of drugs is not always used as first option in empirical treatment for early nosocomial infection in developing countries. Inadequate restriction of carbapenems by some infection control programs may be associated with increasing mortality.

Considering the problem of inadequate treatment and mortality rate among patients with ESBL and non-ESBL strains of K. pneumoniae bacteremia, this study evaluated risk factors for mortality in patients with bloodstream infection caused by ESBL-KP in a Brazilian hospital with high incidence of this microorganism.

METHODS

A retrospective cohort was carried out at the Hospital Universitário Evangélico de Curitiba, a reference center for trauma, burn and renal transplant. This center is a 660-bed tertiary-care hospital in Curitiba, a city located in Southern Brazil with 60 intensive care beds.

From January 2006 to January 2009, 884 positive blood cultures were identified (excluding coagulase negative Staphylococcus) and 11.7% were positive for K. pneumoniae. All the patients with bacteremia caused by K. pneumoniae were included in our study. Patients with more than one positive blood culture were included only once. Only data associated with the first bacteremia and patients older than 12 years-old were evaluated.

Cultures were collected and processed using the BACT/Alert® (bioMérieux, Durham, USA). Klebsiella species were identified using biochemical analysis.3 Susceptibility testing was performed by the disk diffusion method according to the CLSI guidelines.4

The following variables were evaluated for each patient: sex; age; previous hospital admission within 90 days; admission to the intensive care unit; length of hospitalization before bacteremia; use of mechanical ventilation, central venous line, urinary catheter and surgery during the current hospitalization; underlying conditions such as diabetes mellitus, chronic renal failure, heart failure, neoplasm, acute renal failure, trauma and previous antibiotic use during current hospitalization. Previous infection or colonization with K. pneumoniae was evaluated. We assumed that the patients who were not cultured were negative for statistical analyses purposes.

Dynamic variables were included, as heart and respiratory rate, temperature, mean arterial pressure and Glasgow Coma Scale (GCS) in the day of the bacteremia. The following laboratory results were evaluated on the day of bacteremia: hemoglobin level, leukocyte and platelet counts, sodium, potassium, creatinine, urea, total billirubin and partial pressure of oxygen from arterial blood.

Thirty-day or in-hospital mortality was registered. Antibiotic treatment was classified as adequate or inadequate. Treatment of each patient was considered adequate if K. pneumoniae was susceptible to the antibiotic used during bacteremia and treatment was initiated within the first 48 hours of bacteremia diagnosis. Carbapenem (ertapenem, imipenem or meropenem) or polymyxin was considered the adequate antibiotic for ESBL-KP, when correct dose was administered. Tygecycline was not used and the hospital strains were not susceptible to aminoglycoside or fluoroquinolones during the period analyzed. The treatment for non-ESBL-KP was considered adequate if susceptible in vitro.

Statistical analysis

Patients with ESBL-KP bacteremia were compared with those with non-ESBL-KP bacteremia to determine factors associated with carbapenem resistance. A second analysis was performed comparing patients who died during hospitalization with those who survived. Continuous data were expressed as mean ± standard deviation (SD) or median with ranges. Frequencies were expressed as percentages. Dichotomous variables were compared using Chi-square and Mann Whitney test was used for continuous variables. Significance level was set at 0.05. Variables with p < 0.10 in the univariated analysis were included in the multivariable analysis. Multivariate analysis was performed using binary logistic regression model. Odd ratios (OR) with 95% confidence intervals (95% CI) were calculated for each variable. Variables in which 95% CI did not include 1.0 were maintained in the final model.

Kaplan-Meier survival estimates were calculated to evaluate the role of adequate treatment in the outcome of bacteremia caused by ESBL-KP, and the difference was assessed using the log-rank test.

All data were stored using the software Excel (Microsoft, New York, USA) and statistical analysis was performed using the software SPSS 11.5 (SPSS, Chicago, USA). Kaplan-Meier survival estimates were determined with GraphPad Prism 4.0 (GraphPad, La Jolla, USA).

RESULTS

One hundred and four patients were included in this study. Sixty-three patients were enrolled as ESBL-KP (60%) and 41 patients with non-ESBL-KP (40%). The median age of all patients included was 43.0 years (range, 12-86) and 43% were female. The mean length of hospitalization was 45.87 ± 36.60 days and the mean duration of hospitalization until first K. pneumoniae bacteremia was 18.8 ± 17.81 days. Characteristics of the patients and laboratory data are described in the Table 1.

In the univariate analysis, factors associated with ESBL included invasive procedures as mechanical ventilation (p = 0.007), the use of central venous catheter (p = 0.009), and elective surgery (p = 0.023). Other risk factors associated with ESBL were male sex (p = 0.027), diabetes mellitus (p = 0.043), and admission to the intensive care unit (p = 0.017). Previous use of fourth generation cephalosporin (p = 0.003) was a risk factor as well. Although resistant strains were associated to longer stay in the hospital, hospitalization length until the bacteremia was not a risk factor for ESBL infection (Figure 1). Even though these finding supports the association with ESBL infection, only mechanical ventilation (p = 0.022) and central venous line (p = 0.017) turned out to be independent risk factors for ESBL-KP bacteremia in the multivariate analysis. All other investigated risk factors were significant only in the univariate analysis.

Global ESBL-KP and non-ESBL-KP mortality rates were similar (49.2% vs. 41.9%, p = 0.43). Kaplan-Meier estimates of survival were also similar for ESBL-KP and non-ESBL-KP infected patients (Figure 2A). There was no independent risk factor associated with higher mortality rate at the multivariate analysis.

Only 55 patients (52.9%) received adequate treatment within 48 hours of bacteremia. Mortality of patients receiving adequate treatment was 50.9% (28 of 55 patients), and 40.8% in patients with inadequate antibiotic therapy (20 of 49 patients). Mortality of Klebsiella bacteremia was not attributed to inadequate treatment as demonstrated in the Kaplan-Meier curve (Figure 2B). Patients infected by either ESBL or non-ESBL strains who received adequate therapy had similar mortality rates (p = 0.43).

DISCUSSION

In our study, 58% of the Klebsiella strains were ESBL-producing. Marra et al.5 found similar rate, showing 52% of ESBL-KP in another Brazilian hospital. Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) study described a different panorama in North America, Asia and Europe (Northern and Southern) in the SMART study with only 17% ESBL-producers Klebsiella spp.6 Data from the MYSTIC study from South America showed 52% of resistance to cefepime, which can be a surrogate for ESBL strains.7

Use of beta-lactam antibiotics containing an oxyimino group (cefuroxime, ceftriaxone, cefotaxime, ceftazidime, oraztreonam) are risk factors for ESBL-KP.8-11 Nevertheless, previous use of cefepime was a risk factor for ESBL in our study. This risk factor had not been confirmed previously.12,13 In our hospital, cefepime is commonly used for the treatment of severe community pneumonia at the emergency room. The use of cefepime is a modifiable risk factor, and restriction of cefepime use could be an approach to decrease ESBL-KP in our institution. Other risk factors for ESBL strains were invasive procedures, which included central venous line and mechanical ventilation.

Previous studies showed a mortality rate between 20 and 40% in patients with bacteremia caused by K. pneumoniae.14-16 The mortality rate in our study was high, when compared to other studies. Although no direct comparison to patients from other studies can be made, in our study patients were severely ill. The first explanation would be a higher proportion of patients with K. pneumoniae bacteremia receiving inadequate empirical therapy. Nonetheless, despite being elevated the rate of inadequate treatment was not associated with mortality of ESBL bacteremia patients. Du et al.17 demonstrated that inadequate treatment was an isolated risk factor, justifying early and empirical use of carbapenems in patients with bacteremia in some hospitals. The cost of carbapenem restricts its use in developing countries, increasing mortality as demonstrated in Nigeria.16 A previous publication of Paterson18 underscored that early treatment with carbapenems is particularly important in seriously ill patients with nosocomial infections, mainly those caused by ESBL strains, decreasing the mortality rate. However, larger sample size of patients would be required to draw more definitive conclusions.

Some caveats must be considered in our study. It is a Days retrospective analysis from an isolated institution with particular characteristics, the infection control program includes strict restriction of carbapenems, the attributed

mortality for ESBL-KP was not calculated and some variables were not included in the analysis, such as APACHE score. A clonal study should be performed to determine the strain as well as the type of the ESBL bacteria. One of the factors associated with ESBL-KP bacteremia was previous use of cefepime. This could be due to bias as ESBL-KP were compared with non-ESBL producing KP and this group of patients very probably did not use cefepime (as cefepime would have eradicated the susceptible microorganism).

CONCLUSION

In conclusion, mortality due to K. pneumoniae infection is high and more than 50% of patients were under inadequate treatment in the ESBL group, justifying the use of carbapenem as first choice in patients admitted to centers with high incidence of ESBL-producing species. We believe that a change in the empirical treatment is necessary, mainly in patients with sepsis during hospitalization, which can decrease the global mortality of patients infected by ESBL-KP. The time to use this group of drugs should not be guided by hospitalization length, because we could not correlate increased resistance with duration of hospitalization. Our institution will change the carbapenem restriction and assess the long term benefits of this approach in the mortality of patients with ESBL infection as well as the effect on resistance profile. We believe that this approach must be implemented in other institutions according to local epidemiology. Although ESBL constituted < 10% of all bacteremia in our hospital, we can not neglect the 50% inadequate therapy for this group of bacteria.

Submitted on: 11/19/2010

Approved on: 12/20/2010

Conflict of interest: Felipe F. Tuon receives grants from MSD, Pfizer, Novartis and United medical.

This study was supported by Merck (IISP 37333).

©2011 Elsevier Editora Ltda. All rights reserved.

  • 1. Girao E, Levin AS, Basso M, et al. Trends and outcome of 1121 nosocomial bloodstream infections in intensive care units in a Brazilian hospital, 1999-2003. Int J Infect Dis. 2008; 12:e145-e146.
  • 2. Sader HS, Jones RN, Andrade-Baiocchi S, et al. Four-year evaluation of frequency of occurrence and antimicrobial susceptibility patterns of bacteria from bloodstream infections in Latin American medical centers. Diagn Microbiol Infect Dis. 2002; 44:273-80.
  • 3. York, M. 2004. Aerobic Bacteriology. In: H. I. Isenberg (Ed.), Clinical Microbiology Procedures Handbook: 3.17.1-3.17.48. New York: ASM Press.
  • 4. Clinical and Laboratory Standard Institute. Performance standards for antimicrobial disk susceptibility testing; Fifteenth Informational Supplement. Clinical and Laboratory Standard Institute 2005;Wayne.
  • 5. Marra AR, Pereira CA, Castelo A, et al. Health and economic outcomes of the detection of Klebsiella pneumoniae-produced extended-spectrum beta-lactamase (ESBL) in a hospital with high prevalence of this infection. Int J Infect Dis. 2006; 10:56-60.
  • 6. Rossi F, Baquero F, Hsueh PR, et al. In vitro susceptibilities of aerobic and facultatively anaerobic Gram-negative bacilli isolated from patients with intra-abdominal infections worldwide: 2004 results from SMART (Study for Monitoring Antimicrobial Resistance Trends). J Antimicrob Chemother. 2006; 58:205-10.
  • 7. Kiffer C, Hsiung A, Oplustil C, et al. Antimicrobial susceptibility of Gram-negative bacteria in Brazilian hospitals: the MYS-TIC Program Brazil 2003. Braz J Infect Dis. 2005; 9:216-24.
  • 8. Paterson DL, Ko WC, Von GA, et al. International prospective study of Klebsiella pneumoniae bacteremia: implications of extended-spectrum beta-lactamase production in nosocomial Infections. Ann Intern Med. 2004; 140:26-32.
  • 9. Urbanek K, Kolar M, Loveckova Y, et al. Influence of thirdgeneration cephalosporin utilization on the occurrence of ESBL-positive Klebsiella pneumoniae strains. J Clin Pharm Ther. 2007; 32:403-8.
  • 10. Hyle EP, Bilker WB, Gasink LB, et al. Impact of different methods for describing the extent of prior antibiotic exposure on the association between antibiotic use and antibiotic-resistant infection. Infect Control Hosp Epidemiol. 2007; 28:647-54.
  • 11. Calbo E, Romani V, Xercavins M, et al. Risk factors for community-onset urinary tract infections due to Escherichia coli harbouring extended-spectrum beta-lactamases. J Antimicrob Chemother. 2006; 57:780-3.
  • 12. Paterson DL, Ko WC, Von GA, et al. International prospective study of Klebsiella pneumoniae bacteremia: implications of extended-spectrum beta-lactamase production in nosocomial Infections. Ann Intern Med. 2004; 140:26-32.
  • 13. Marra AR, Wey SB, Castelo A, et al. Nosocomial bloodstream infections caused by Klebsiella pneumoniae: impact of extended-spectrum beta-lactamase (ESBL) production on clinical outcome in a hospital with high ESBL prevalence. BMC Infect Dis. 2006; 6:24.
  • 14. Trecarichi EM, Tumbarello M, Spanu T, et al. Incidence and clinical impact of extended-spectrum-beta-lactamase (ESBL) production and fluoroquinolone resistance in bloodstream infections caused by Escherichia coli in patients with hematological malignancies. J Infect. 2009; 58:299-307.
  • 15. Memon JI, Rehmani RS, Ahmed MU, et al. Extended spectrum beta-lactamase producing Escherichia coli and Klebsiella pneumoniae bacteremia. Risk factors and outcome in the eastern region of Saudi Arabia. Saudi Med J. 2009; 30:803-8.
  • 16. Menashe G, Borer A, Yagupsky P, et al. Clinical significance and impact on mortality of extended-spectrum beta lactamase-producing Enterobacteriaceae isolates in nosocomial bacteremia. Scand J Infect Dis. 2001; 33:188-93.
  • 17. Du B, Long Y, Liu H, et al. Extended-spectrum beta-lactamaseproducing Escherichia coli and Klebsiella pneumoniae bloodstream infection: risk factors and clinical outcome. Intensive Care Med. 2002; 28:1718-23.
  • 18. Paterson DL, Ko WC, Von GA, et al. Antibiotic therapy for Klebsiella pneumoniae bacteremia: implications of production of extended-spectrum beta-lactamases. Clin Infect Dis. 2004; 39:31-7.

  • Correspondence to:
    Felipe Francisco Tuon
    Alameda Augusto Stellfeld, 1908, 4º andar, CCIH - Bigorrilho
    80730-150, Curitiba, PR, Brazil
    Phone: 55 41-3240-5000

Publication Dates

  • Publication in this collection
    04 Jan 2012
  • Date of issue
    Dec 2011

History

  • Received
    19 Nov 2010
  • Accepted
    20 Dec 2010
Brazilian Society of Infectious Diseases Rua Augusto Viana, SN, 6º., 40110-060 Salvador - Bahia - Brazil, Telefax: (55 71) 3283-8172, Fax: (55 71) 3247-2756 - Salvador - BA - Brazil
E-mail: bjid@bjid.org.br