Malaysian Journal of Medical Sciences School of Medical Sciences, Universiti Sains Malaysia ISSN: 1394-195X Vol. 14, Num. 2, 2007, pp. 36-40

ORIGINAL ARTICLE

Hospital Outcomes of Adult Respiratory Tract Infections with Extended-Spectrum B-Lactamase (ESBL) Producing Klebsiella Pneumoniae

Li-Cher Loh, Nor Izran Hanim bt Abdul Samad, Rosdara Masayuni bt Mohd Sani, Sree Raman*, Tarmizi Thayaparan*, Shalini Kumar**

IMU Lung Research, International Medical University, Kuala Lumpur, Malaysia; *Department of Medicine, Seremban Hospital, Negeri Sembilan, Malaysia; **Department of Pathology, Seremban Hospital, Negeri Sembilan, Malaysia
Corresponding Author : Dr Li-Cher Loh MBBCh (Ireland), MRCP (UK), MD (London), Department of Medicine, Clinical School, International Medical University, Jalan Rasah, Seremban 70300, Negeri Sembilan, Malaysia Tel: (+606) 767 7798 Fax: (+606) 767 7709 E mail: richardLloh@imu.edu.my

Submitted-20-02-2005, Accepted-03-12-06

Code Number: mj07023

Klebsiella pneumoniae ranks high as a cause of adult pneumonia requiring hospitalization in Malaysia. To study whether extended-spectrum b-lactamase (ESBL) producing K. pneumoniae was linked to hospital outcomes, we retrospectively studied 441 cases of adult respiratory tract infections with microbial proven K. pneumoniae from an urban-based university teaching hospital between 2003 and 2004. 47 (10.6%) cases had ESBL. Requirement for ventilation and median length of hospital stay, were greater in ‘ESBL’ than in ‘non-ESBL’ group [34% vs. 7.4%, p<0.001; 14 days vs. 5 days, p<0.001 respectively] but not crude hospital mortality rate [21.3% vs. 12.4%, p=0.092]. There was a four-fold increased risk of requiring ventilation [4.61 (2.72-7.85)] when ESBL was present. Our findings support the association of ESBL producing K. pneumoniae with adversed hospital outcomes and reiterate the need for vigilance on the part of treating clinicians.

Key words : extended spectrum beta-lactamase, ESBL, Klebsiella pneumoniae, hospitalization, respiratory tract infections, mortality, Malaysia

Introduction

Klebsiella pneumoniae, the most clinically important of all Kebsiella species (1), ranks high as a cause of community-acquired pneumonia (CAP) in adult hospitalized patients in Malaysia (2,3,4). This appears unique as neighbouring countries in the region like Thailand (5) and Singapore (6) do not always share this finding.

Extended-spectrum b-lactamases (ESBLs) is a problematic resistance mechanism that is commonly associated with K. pneumoniae. Since its first recognition in the United States in the 1980s, it is now reported worldwide (7) including in Malaysia (8). It poses important therapeutic dilemma in that most b-lactam stable antibiotics are ineffective towards ESBL producing organisms. Third-generation cephalosporins, in particular, have variable therapeutic efficacies (9) and are shown to have high propensity to induce the development of such resistance mechanism (10). Presence of ESBL producing K. pneumoniae infection has been shown to be associated with adverse clinical outcomes (11,12) and in the treatment of any pneumonia, appropriate choice of empiric antibiotics is important and has been shown to affect outcomes (13,14).

In view of the prevalence of K. pneumoniae in hospitalized adult pneumonia cases in Malaysia and the lack of published data on the clinical impact of ESBL-producing K. pneumoniae in the context of respiratory tract infections, we carried out a retrospective observational study of patients with K. pneumoniae respiratory tract infections admitted to our university teaching hospital, to find out whether respiratory tract infections caused by ESBL producing K. pneumoniae affected clinical outcomes in terms of hospital mortality, requirement for mechanical ventilation and length of hospital stay.

Methods & Patients

Data collection

Data on all respiratory specimens (i.e. sputum, tracheal aspirates, bronchial washing and bronchoalveolar lavage) that cultured K. pneumoniae between January 2003 and December 2004 from our 800-bed urban-based university teaching hospital and their patient details were downloaded from hospital computer. Medical records of these patients were then retrieved and reviewed for eligible cases. Included were adult patients (≥12 years of age) who received antibiotic treatment for respiratory tract infections diagnosed on clinical (plus or minus) radiological grounds and in whom respiratory isolates cultured K. pneumoniae. For the purpose of this study, we did not seek to make any distinction between community- and hospital-acquired cases.

Data was collected using a standardized form and consisted of individual patient, clinicodemographic details, antibiotic in vitro susceptibility results, and clinical outcomes in terms of crude hospital mortality rate, requirement of mechanical ventilation and length of hospital stay. Co-morbidity scoring (15) was carried out by an investigator (LCL), without the knowledge of patient groups. The study protocol was approved by local university research and ethics committee (International Medical University Research and Ethics Committee, number 065/2004).

Klebsiella Pneumoniae

K. pneumoniae was identified by standard microbiological culture technique to the level of species. All enterobacteriaceae cultured were tested for susceptibility to a panel of 6 to 9 antibiotics prespecified by the Ministry of Health Malaysia (Personal communication, Hj Abd Jalil Mohd, Microbiology Unit), and tested for extendedspectrum b-lactamases, using a double-disc synergy test.

Statistical analysis

For analysis, patients were divided into two groups according to presence of ESBL. Differences between groups were assessed by Chi-Square tests for categorical, and t tests or Mann Whitney tests depending on the normality of continuous data. Odd ratios with 95% confidence interval were calculated to determine their relative risks. All computation was made using statistical package SPSS version 11.5 for Windows (Chicago, Illinois, USA). In all cases, statistical significance was defined at the 5% level and assessed based with two-tailed tests.

Results

Of the 441 eligible patients with documented respiratory tract infections and K. pneumoniae cultured from respiratory specimens, 47 (10.6%) had ESBL-producing strains. The clinico-demographic features are described in Table 1. Both groups were comparable except for type of respiratory specimens from which the organisms were isolated.

The overall rates for hospital mortality and requirement for mechanical ventilation of patients with K. pneumoniae were 13.4% and 10.2%. The mortality rate during hospital stay of patients with ESBL and non-ESBL producing K. pneumoniae were 21.3% and 12.4% respectively (p=0.092). The percentage of patients requiring mechanical ventilation were 34% and 7.4% respectively (p<0.001). The median length of hospital stay in patients with and without ESBL producing K. pneumoniae were 14 (Interquartile range^25-75, IQR^25-75: 8-32) and 5 (IQR^25-75: 3-10) days respectively (p<0.001) [Figure 1]. There was a four-fold increased risk of requiring ventilation [4.61 (2.727.85)] when ESBL was present.

Discussion

We have shown that in adult patients with respiratory tract infections ESBL producing K. pneumoniae, is associated with more adversed hospital outcomes in relation to the need for mechanical ventilation and having extended hospital stay when compared to non-ESBL.

The association between ESBL producing K. pneumoniae and adversed clinical outcomes is well documented. In an international prospective study of 455 bacteraemic patients (16), Paterson et al showed that ESBL-producing K. pneumoniae was associated with high mortality with failure of correct antibiotics. In another recent study of 133 ESBL producing K. pneumoniae and Escherichia Coli bacteraemic patients (17), Kang CI et al showed that appropriate antibiotics, not necessarily those that were empirically instituted, was clinically important. In fact, they showed that delay in appropriate definitive antimicrobial therapy was not associated with higher mortality if antimicrobial therapy was adjusted appropriately according to the susceptibility results. However, very few studies to date have investigated clinical outcomes in the context of pneumonia. Our results add to the existing literature, the evidence of association between ESBLproducing organisms and adversed clinical outcomes in patients with pneumonia.

Apart from the inherent weaknesses of a retrospective study, our study protocol had not included other clinical parameters (18), such as oxygenation status, blood urea, the presence of confusion, radiological extent of pneumonia, in the assessment of hospital clinical outcomes between ESBL and non-ESBL groups. Nevertheless, our data have taken into account the two key parameters of age and co-morbidity in severity assessment of any pneumonia. Furthermore while ESBL cases our overall sample size of K. pneumoniae cases is relatively large. Also it noteworthy, that our study did not make any distinction between pneumonia and respiratory tract infections because the retrospective nature of the study could not allow us to be confident in such distinction. We are however confident from our data that these were patients treated for respiratory tract infections.

Another important concern is whether ESBLs isolated from tracheal aspirates included in our analysis would inherently bias the findings towards adversed hospital outcomes since they represent more critically ill patients on ventilators. In a separate further analysis using sputum cases alone, we could show that the findings were the same i.e. ESBL group had a significantly higher rate of mechanical ventilation (13.3% vs. 2.8%; p=0.03) and longer hospital stay (median days: 15 vs. 6; p<0.001). In view of this, we are confident that our conclusions have not been significantly altered by this factor.

Although the prevalence of K. pneumoniae in clinical respiratory isolates of our hospital has increased over the years (unpublished data), the prevalence of ESBL producing K. pneumoniae remains low (<15%). In this study, about 10% of K. pneumoniae cases were ESBL producing. Such findings like ours provide local benchmark data for the country and the region. However the local epidemiology data are likely to vary from one setting to another. A study (18) of all 570 clinical isolates from four medical centres in Malaysia and two medical centres in Singapore showed a prevalence of ESBL producing K. pneumoniae of between 36.7% and 38%. It is obvious that continuing national and regional surveillance and research will be continually needed to monitor patterns in order to develop appropriate healthcare policies and control (19).

Acknowledgements

This study is funded by International Medical University internal research fund (IMU 065/2004). The authors wish to thank Mr Haji Abdul Latiff Mohamed, Senior Microbiologist, Microbiology Unit, Department of Pathology, Seremban Hospital, and staff in medical records of Seremban Hospital, for their kind assistance in retrieving the relevant data.

Reference

1. Toner, M., and Irimia, D. Blood-on-a-chip Annual Review of Biomedical Engineering 2005; 7:77–103.
2. Boyum, A. Isolation of leucocytes from human blood: A two-phase system for removal of red cells with methylcellulose as erythrocytes aggregative agent Scandinavian Journal of Clinical and Laboratory Investigation 1968; 21:9–29.
3. Boyum, A. Isolation of leucocytes from human blood: Further observations – methylcellulose, dextran and Ficoll as erythrocyte aggregating agents Scandinavian Journal of Clinical and Laboratory Investigation 1968; 31:50.
4. Tiirikainen, MI. Evaluation of red blood cell lysing solutions for the detection of intracellular antigens by flow cytometry Cytometry 1995; 20: 341 – 348.
5. Gaudernack, T., Leivestad, T., Ugelstad, J. and Thorsby, E. Isolation of pure functionally active CD8+ T cells. Positive selection with monoclonal antibodies directly conjugated to monosized magnetic microspheres Journal of Immunological Methods 1986; 90:179–187.
6. Olsvik, Ø., Popovic, T., Skjerve, E., Cudjoe, KS., Hornes, E., Ugelstad, J. and Uhlén, M. Magnetic separation techniques in diagnostic microbiology Clinical Microbiology Review 1994; 7:43–54.
7. Molnar, B., Sipos, F., Galamb, O. and Tulassay, Z. Molecular detection of circulating cancer cells Digestive Diseases 2003; 21:320–325.
8. Hayflick, L. and Moorhead, PS. The serial cultivation of human diploid cell strains Experimental Cell Research 1961; 25:585–621.
9. Bodnar, AG., Ouellette, M., Frolkis, M., Holt, SE., Chiu, C-P., Morin, GB., Harley, CB., Shay, JW., Lichsteiner, S. and Wright, WE. Extension of life-span by introduction of telomerase into normal human cells Science 1998; 279:349–352.
10. Mosmann, S. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays Journal of Immunological Methods 1983; 65:55–63.
11. Mizuki E, Ohba M, Akao T, Yamashita S, Saitoh H and Park YS. Unique activity associated with noninsecticidal Bacillus thuringiensis parasporal inclusions: in vitro cell-killing action on human cancer cells Journal of Applied Microbiology 1999; 86:477–486.
12. Mizuki E, Park YS, Saitoh H, Yamashita S, Akao T, Higuchi K and Ohba M. Parasporin, a human leukemic cell-recognizing parasporal protein of Bacillus thuringiensis Clinical Diagnostic and Laboratory Immunology 2000; 7:625–634.
13. Nadarajah, VD., Chai, SH. Diana T, Mohamed, SM, Michelle NYT, Kumaresan S. The bioactivity of Bacillus thuringiensis - diversity in the Malaysian strains Proceedings of the 14th National Biotechnology Seminar 2003; 243–248.
14. Toivanen, P. and Toivanen, A. Selective activation of chicken T lymphocytes by concanavalin A Journal of Immunology 1973; 111: 1602–0606.
15. Loa, CC, Lin, TL. and Wu, CC. Factors affecting mitogenic response of turkey lymphocytes Acta Veterinary Brno 2001; 70: 433–442. [* PICT is empty or cannot be processed. | In-line Graphic *]

© Copyright 2007 - Malaysian Journal of Medical Science

The following images related to this document are available:

Photo images