Journal of Hospital Infection 77 (2011) 134e137

Available online at www.sciencedirect.com

Journal of Hospital Infection

journal homepage: www.elsevierhealth.com/journals/jhin

Risk factors for multidrug-resistant bacterial infection among patients with

tuberculosis
H.-R. Kim a,1, S.S. Hwang b, E.-C. Kim c, S.M. Lee a, S.-C. Yang a, C.-G. Yoo a, Y.W. Kim a, S.K. Han a,
Y.-S. Shim a, J.-J. Yim a, *
a
Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Lung Institute of Medical Research Center, Seoul National University College of Medicine,
Seoul, Republic of Korea
b
Department of Social and Preventive Medicine, College of Medicine, Inha University, Incheon, Republic of Korea
c
Department of Laboratory Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea

a r t i c l e i n f o s u m m a r y

Article history: Given that anti-tuberculosis medication itself has antibacterial activity and that broad-spec-
Received 27 September 2009
trum antibiotics are frequently used, the emergence of multidrug-resistant (MDR) bacteria
Accepted 2 July 2010
Available online 17 September 2010 among patients being treated for tuberculosis (TB) is likely. We used a caseecontrol design to
study the clinical predictors of MDR bacterial infection among TB patients. Both cases and
Keywords: controls were selected from among patients who were diagnosed and treated as having TB
Bacterial infection between 1 January 1996 and 31 August 2006. TB patients with MDR bacterial infection were
Drug resistance included as cases and those with non-MDR bacterial infection were included as controls.
Tuberculosis
Multiple logistic regression analysis was performed to elucidate the risk factors for MDR
bacterial infection. During the study period 3667 patients were diagnosed with, and treated
for, TB. A total of 123 experienced episodes of bacterial infection, of whom 59 (48.0%) were
infected by an MDR strain at least once. The presence of chronic renal failure [adjusted odds
ratio (OR): 4.96; 95% confidence interval (CI): 1.37e18.01] and the use of antimicrobials other
than typical anti-TB drugs within three months (adjusted OR: 4.37; 95% CI: 1.74e10.95) were
independent risk factors for MDR bacterial infection. Bacterial infection in TB patients is
commonly multidrug resistant. Clinicians should be aware of the possibility of MDR bacterial
infection among TB patients with chronic renal failure or recent use of other antimicrobials.
Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.

Introduction infection caused by multidrug-resistant (MDR) bacteria, is

The introduction of each new class of antimicrobial agent has
been followed by the emergence of resistance to that class over
time, rendering hospitalised patients vulnerable to infection that is
not treated effectively by routinely used agents or by any available
agent.1 The development of antimicrobial resistance is associated
with increases in mortality, morbidity, length of hospitalisation,
and cost of healthcare.2 Unfortunately, infection due to bacteria
with antimicrobial resistance, especially healthcare-associated
growing.3 The known risk factors for infection by MDR bacteria
include advanced age, underlying disease, prolonged hospital-
isation, gastrointestinal surgery or transplantation, the use of
invasive devices, and exposure to antimicrobial drugs.4
Given that anti-tuberculosis (TB) medication (e.g. rifampicin)
itself has antibacterial activity and that broad-spectrum antibiotics
such as fluoroquinolones and aminoglycosides are frequently used in
patients with hepatotoxicity or MDR-TB, the emergence of multidrug
resistance is likely in TB patients.5e7 In this study we examined the
clinical predictors of MDR bacterial infection among patients with TB.

* Corresponding author. Address: Division of Pulmonary and Critical Care Medi- Methods
cine, Department of Internal Medicine, Lung Institute of Medical Research Center,
Seoul National University College of Medicine, 101 Daehangno, Jongno-gu, Seoul 110- Study design and subjects
744, Republic of Korea. Tel.: þ82 2 2072 2059; fax: þ82 2 2072 9662.
E-mail address: yimjj@snu.ac.kr (J.-J. Yim).
1
Present address: Department of Internal Medicine, Korea Cancer Center We used a caseecontrol design. Both cases and controls were
Hospital, 215-4, Gongneung-dong, Nowon-gu, Seoul, 139-706, Republic of Korea. selected from among patients who were diagnosed and treated as

0195-6701/$ e see front matter Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jhin.2010.07.004
 H.-R. Kim et al. / Journal of Hospital Infection 77 (2011) 134e137 135

having TB between 1 January 1996 and 31 August 2006 at Seoul Table I

National University Hospital, a tertiary university-affiliated hospital in Demographic and clinical characteristics of the 123 tuberculosis (TB) patients with
episodes of bacterial infection
South Korea. All patients from whom bacteria other than Mycobac-
terium tuberculosis were isolated during anti-TB treatment or within Characteristics Total (N ¼ 123)
six months of completion of treatment were included. Patients with Age (years), median (range) 58 (18e93)
MDR bacterial infection were included as cases, and patients Male sex 66 (53.7%)
Body mass index (kg/m2) 20.9  3.8
with non-MDR bacterial infection were included as controls. Patients
Current or ex-smoker 41 (33.3%)
with more than one episode of infection that included at least Comorbidities
one episode caused by MDR bacteria were classified as cases. The Diabetes 34 (27.6%)
protocol for this study was approved by the ethics review committee Chronic liver diseasea 12 (9.8%)
of Seoul National University Hospital. Several previous studies by our Chronic renal failureb 18 (14.6%)
Malignancy 24 (19.5%)
group have been based on this population of TB patients.8e18 HIV seropositivity 3 (2.4%)
Immunosuppressant therapy 27 (22.0%)
Review of clinical records Residence in long term care facility 4 (3.3%)
Characteristics of TB treatment
No. of used anti-TB drugs, median (range) 4 (2e10)
Patient demographics (sex, age, body mass index, and smoking
MDR-TB 12 (9.8%)
status) and clinical characteristics (comorbidities, types of TB, and Status of anti-TB treatment during infection episode
anti-TB drugs used) were obtained from clinical records. In partic- During anti-TB treatment 96 (78.0%)
ular, the following risk factors for drug-resistant bacterial infection After completion of anti-TB treatment 27 (22.0%)
were sought: the use of antimicrobials other than typical anti-TB Duration of anti-TB treatment before bacterial 5 (1e63)
culture, median (range), months
drugs (isoniazid, rifampicin, pyrazinamide, ethambutol, cyclo- Risk factors for bacterial infection by MDR bacteria
serine, para-aminosalicylic acid, prothionamide, and capreomycin); Nosocomial infection 82 (66.7%)
nosocomial infection; residence in a long term care facility; recent Use of antimicrobials other than typical anti-TB drugs 59 (48.0%)
admission; mechanical ventilation; indwelling central venous within 3 months
Recent admission (within 3 months) 84 (68.3%)
catheterisation; and urinary bladder catheterisation.
Recent admission to intensive care unit within 3 months 25 (20.3%)
Mechanical ventilation 9 (7.3%)
Definitions Indwelling central venous catheterisation 16 (13.0%)
Urinary bladder catheterisation 21 (17.1%)
Episode of infection HIV, human immunodeficiency virus; MDR, multidrug resistant.
a
Infection was defined using the criteria proposed by the Centers Chronic liver disease includes hepatitis, alcoholic liver disease, cirrhosis, liver
for Disease Control and Prevention (CDC, Atlanta, GA, USA).19 failure.
b
Chronic renal failure is defined as a decreased kidney glomerular filtration rate
Episodes of infection were excluded if no organism was isolated
<60 mL/min/1.73 m2 for 3 months.
from cultures of tissue or fluid from the affected site. Infections
were categorised based on anatomical sites: surgical wound controls. Analyses were performed using Pearson’s c2-test or
infection; primary bloodstream infection; respiratory system Fisher’s exact test for categorical variables, and Student’s t-test for
infection; urinary tract infection; bone and joint infection; cardio- continuous variables. Variables with P < 0.20 in univariate
vascular system infection; central nervous system infection; eye, comparisons were subjected to multiple logistic regression to
ear, nose, throat, and mouth infection; gastrointestinal system identify predictors of MDR bacterial infection. In logistic regression,
infection; reproductive tract infection; skin and soft tissue infec- backward elimination was used to select variables to be maintained
tion; and systemic infection. Infection developing after 48 h of in the final model, using P < 0.10 as the criterion for statistical
hospitalisation was classified as nosocomial infection. significance of associations. The area under the receiver operator
characteristic (ROC) curve was used to evaluate the performance of
Multidrug-resistant (MDR) bacteria the models. All statistical analyses were performed with Stata
MDR bacteria included meticillin-resistant Staphylococcus software (version 10.0; StataCorp., College Station, TX, USA).
aureus; meticillin-resistant coagulase-negative staphylococci;
vancomycin-resistant enterococci; Streptococcus pneumoniae
Results
resistant to penicillin and other broad-spectrum agents, such as
macrolides and fluoroquinolones; and MDR Gram-negative
Characteristics of the 123 TB patients with episodes of infection
bacilli.20 MDR Gram-negative bacilli included Klebsiella spp.,
Enterobacter cloacae and Escherichia coli resistant to at least three of
During the study period, 3667 patients were diagnosed and
the following antimicrobial groups: third or fourth generation
treated at Seoul National University Hospital as having TB. From
cephalosporins, aminoglycosides, fluoroquinolones, piperacillin,
and ampicillin/sulbactam. Pseudomonas aeruginosa resistant to Table II
at least three of the following antimicrobial groups was also clas- Anatomical sites of bacterial infection in 321 episodes from the 123 tuberculosis
sified as an MDR Gram-negative bacillus: ceftazidime/cefepime, patients

aminoglycosides, fluoroquinolones, carbapenems, and piperacillin. Anatomical sites Total (N ¼ 321)

Acinetobacter baumannii resistant to all antimicrobial agents, or Urinary tract 111 (34.6%)
all except imipenem, and organisms such as Stenotrophomonas Respiratory system 94 (29.2%)
maltophilia that are intrinsically resistant to the broadest spectrum Skin/soft tissue 35 (10.9%)
Gastrointestinal system 31 (9.7%)
antimicrobial agents were also regarded as MDR bacteria.20,21
Primary bloodstream 18 (5.6%)
Surgical wound 16 (5.0%)
Statistical analysis Bone/joint 9 (2.8%)
Eye, ear, nose, throat and mouth 3 (0.9%)
The analysis was based on the initial episodes of MDR bacterial Central nervous system 3 (0.9%)
Cardiovascular system 1 (0.3%)
infection among cases and initial episodes of infection among
136 H.-R. Kim et al. / Journal of Hospital Infection 77 (2011) 134e137

Table III infection. Fifty-nine patients (48.0%) had at least one episode of
Bacteria causing 321 infection episodes among 123 tuberculosis patients MDR bacterial infection.
Organisms Total No. of MDR The median age of the 123 TB patients with infectious episodes
(N ¼ 321) strainsa was 58 years; 66 of them were male. Twelve (9.8%) had MDR-TB. In
Gram-positive bacteria 141 (43.9%) 64 (45.4%) addition, 84 (68.3%) patients had a history of admission within
Enterococci 56 (17.4%) 6 (10.7%) three months. In 59 patients (48.0%), antimicrobials other than
Coagulase-negative staphylococci 42 (13.1%) 32 (76.2%)
typical anti-TB drugs were used for bacterial infection, MDR-TB, or
Staphylococcus aureus 27 (8.4%) 26 (96.3%)
Streptococci other than Streptococcus pneumoniae 8 (2.5%) 0 other causes for at least one week within three months of the
Clostridium difficile 5 (1.6%) 0 infectious episode (Table I).
Streptococcus pneumoniae 2 (0.6%) 0
Othersb 1 (0.3%) 0
Gram-negative bacteria 180 (56.1%) 56 (31.1%) Characteristics of the 321 episodes of infection
Pseudomonas aeruginosa 43 (13.4%) 18 (41.9%)
Escherichia coli 38 (11.8%) 9 (23.7%) The most common anatomical site of infection was the urinary
Klebsiella spp. 20 (6.2%) 6 (30.0%)
tract (111 cases, 34.6%), followed by the respiratory system
Acinetobacter baumannii 18 (5.6%) 11 (61.1%)
Enterobacter cloacae 15 (4.7%) 3 (20.0%) (94 cases, 29.2%; Table II). Gram-positive bacteria accounted for
Stenotrophomonas maltophilia 9 (2.8%) 9 (100%) 141 episodes and Gram-negative bacteria for 180 (Table III).
Othersc 37 (11.5%) 0

MDR, multidrug resistant.
a
Percentages represent the proportion of MDR strains among individual bacteria.
b
Bacillus sp. (1 case).
c
Proteus mirabilis (1 case), Aeromonas caviae (1 case), Bacteroides merdae (1 case),
Burkholderia cepacia (8 cases), Citrobacter freundii (6 cases), Enterobacter aerogenes
(3 cases), Enterobacter amnigenus (1 case), Enterobacter asburiae (1 case), Escherichia
hermanii (1 case), Haemophilus influenzae (1 case), Moraxella catarrhalis (1 case),
Morganella morganii (1 case), Pantoea spp.. (1 case), Prevotella oralis (1 case), Pseu-
domonas putida (1 case), Salmonella spp.. (2 cases), Serratia marcescens (6 cases).
685 of these patients 3200 bacterial cultures were performed
during treatment or within six months of its completion. Ulti-
mately, 932 bacteria other than M. tuberculosis were isolated from
clinical specimens from 211 TB patients. A total of 321 bacterial
isolates in 123 patients (58.3%) met the CDC definition of true
Risk factors for infection by MDR bacteria among the 123 TB
patients
Based on the clinical variables included in the univariate
comparison of the patients with MDR and non-MDR bacterial
infection, the final multiple logistic regression model predicting
MDR bacterial infection included the presence of chronic renal
failure, an episode of infection during anti-TB treatment, and use of
antimicrobials other than typical anti-TB drugs within three
months. Of these, the presence of chronic renal failure [adjusted
odds ratio (OR): 4.96; 95% confidence interval (CI): 1.37e18.01] and
the use of antimicrobials other than typical anti-TB drugs within
three months (adjusted OR: 4.37; 95% CI: 1.74e10.95) were inde-
pendent risk factors (Table IV). The fit of the final model was good
(area under the ROC curve: 0.72; 95% CI: 0.63e0.809).

Table IV
Risk factors of an infection episode by multidrug-resistant (MDR) bacteria among 123 tuberculosis (TB) patients

Variables Bacterial group Unadjusted ORa (95% CI) Adjusted OR (95% CI)

Non-MDR (64 patients) MDR (59 patients)

Age (years), median (range) 58.5 (20e84) 58 (18e93) 0.99 (0.97e1.01)
Male sex 32 (50.0%) 34 (57.6%) 1.36 (0.67e2.77)
Body mass index (kg/m2) 20.7  3.9 21.0  3.7 1.02 (0.91e1.14)
Current or ex-smokerb 21 (35.0%) 20 (35.1%) 1.00 (0.47e2.15)
Comorbidities
Diabetes 19 (30.2%) 15 (25.4%) 0.79 (0.36e1.75)
Chronic liver disease 6 (9.5%) 6 (10.2%) 1.08 (0.33e3.54)
Chronic renal failure 4 (6.3%) 14 (23.7%) 4.59 (1.41e14.89) 4.96 (1.37e18.01)
Malignancy 15 (23.8%) 9 (15.3%) 0.58 (0.23e1.44)
HIV seropositivity 1 (1.6%) 2 (3.4%) 2.18 (0.19e24.64)
Immunosuppressant therapy 11 (17.5%) 16 (27.1%) 1.76 (0.74e4.19)
Characteristics of TB
No. of used anti-TB drugs, median (range) 4 (2e10) 4 (3e9) 1.07 (0.83e1.38)
MDR-TB 8 (12.5%) 4 (6.8%) 0.51 (0.15e1.79)
Status of anti-TB treatment during infection episode
During anti-TB treatment 45 (70.3%) 51 (86.4%) 2.69 (1.08e6.74) 2.83 (0.99e8.07)
After completion of anti-TB treatment 19 (29.7%) 8 (13.6%) 1 (reference)
Risk factors for bacterial infection by MDR strain
Residence of long term care facility 1 (1.6%) 3 (5.1%) 3.38 (0.34e33.38)
Nosocomial infection 36 (56.3%) 46 (78.0%) 2.75 (1.25e6.06)
Admission within previous three months 40 (62.5%) 44 (74.6%) 1.76 (0.81e3.82)
Admission to intensive care unit within 3 months 9 (14.1%) 16 (27.1%) 2.27 (0.92e5.64)
Mechanical ventilation 3 (4.7%) 6 (10.2%) 2.30 (0.55e9.66)
Indwelling central venous catheterisation 5 (7.9%) 11 (18.6%) 2.66 (0.86e8.18)
Urinary bladder catheterisation 8 (12.7%) 13 (22.0%) 1.94 (0.74e5.09)
Use of antimicrobials other than typical anti-TB drugs 33 (53.2%) 48 (84.2%) 4.69 (1.97e11.18) 4.37 (1.74e10.95)
within 3 monthsc

OR, odds ratio; CI, confidence interval.

a
OR for infection by MDR bacteria.
b
Patients with smoking history of at least 1 pack-year.
c
Use of following at least for one week: fluoroquinolones (ofloxacin, ciprofloxacin, levofloxacin, moxifloxacin); aminoglycoside (streptomycin, kanamycin, amikacin);
macrolide (azithromycin, clarithromycin); amoxicillin clavulanate, cephalosporins, clindamycin, metronidazole, piperacillin/tazobactam, carbapenems, vancomycin.
 H.-R. Kim et al. / Journal of Hospital Infection 77 (2011) 134e137
Discussion
Among bacteria other than M. tuberculosis causing infection
during anti-TB treatment or within 6 months of its completion, 120
of 321 strains (37.4%) were multidrug resistant. Chronic renal
failure and the use of antimicrobials other than typical TB drugs
within three months were independent risk factors for MDR
bacterial infection. The same risk factors have previously been
identified for MDR bacterial colonisation or infection in other
groups of patients.4,22e24 The frequent isolation of MDR bacteria
among patients with renal failure has been attributed to the higher
exposure to antibiotics during their clinical courses.25
Our results suggest that treatment of TB patients with antimi-
crobials other than typical anti-TB drugs (whether for the treat-
ment of intercurrent bacterial infection, as alternatives to TB drugs
causing toxicity, or for the treatment of MDR-TB) causes a four-fold
increase in the risk that any subsequent bacterial infection will be
caused by an MDR strain. This observation underscores the
importance of judicious use of antimicrobials in TB patients.
MDR-TB per se was not associated with bacterial infection by
MDR strains in our study but, since there were only 12 such
patients, the possibility that the treatment of MDR-TB is a risk factor
for MDR bacterial infection cannot be excluded.
TB patients who adhere to their treatment are regarded as
harmless to the community once culture conversion is achieved.
However, our findings indicate that some TB patients, especially
those who have had antimicrobials other than typical anti-TB
drugs, could be sources of healthcare-associated MDR bacterial
infection.
Our study has two substantial limitations. First, although we
found that bacteria responsible for other infections in TB patients
were commonly multidrug resistant, we did not compare multidrug
resistance in TB patients with that in patients without TB. Without
this comparison, we cannot conclude that bacterial infection in TB
patients is more likely to be caused by MDR strains. Second, the
power of the study to identify risk factors for bacterial resistance is
limited by the fact that only 123 patients with episodes of bacterial
infection were identified in our population of 3667 TB patients.
Bacterial infection in TB patients is commonly caused by MDR
bacteria. The presence of chronic renal failure and use of antimi-
crobials other than typical anti-TB drugs are independent risk
factors. Clinicians should be aware of the possibility of multidrug
resistance in these circumstances.
Acknowledgement
The authors thank Drs H.b. Kim and Y.A. Kang for insightful
advice on this manuscript.
Conflict of interest statement
None declared.
Funding sources
None.
137
References
1. Santos RP, Mayo TW, Siegel JD. Healthcare epidemiology: active surveillance
cultures and contact precautions for control of multidrug-resistant organisms:
ethical considerations. Clin Infect Dis 2008;47:110e116.
2. Cosgrove SE. The relationship between antimicrobial resistance and patient
outcomes: mortality, length of hospital stay, and health care costs. Clin Infect
Dis 2006;42(Suppl. 2):S82e89.
3. McDonald LC. Trends in antimicrobial resistance in health care-associated
pathogens and effect on treatment. Clin Infect Dis 2006;42(Suppl. 2):
S65e71.
4. Safdar N, Maki DG. The commonality of risk factors for nosocomial colonization
and infection with antimicrobial-resistant Staphylococcus aureus, enterococcus,
gram-negative bacilli, Clostridium difficile, and Candida. Ann Intern Med
2002;136:834e844.
5. Brufani M, Cerrini S, Fedeli W, Vaciago A. Rifamycins: an insight into biological
activity based on structural investigations. J Molec Biol 1974;87:409e435.
6. Maggi N, Pasqualucci CR, Ballotta R, Sensi P. Rifampicin: a new orally active
rifamycin. Chemotherapy 1966;11:285e292.
7. Blumberg HM, Burman WJ, Chaisson RE, et al. American Thoracic Society/
Centers for Disease Control and Prevention/Infectious Diseases Society
of America: treatment of tuberculosis. Am J Respir Crit Care Med 2003;167:
603e662.
8. Cho YJ, Lee SM, Yoo CG, et al. Clinical characteristics of tuberculosis in patients
with liver cirrhosis. Respirology 2007;12:401e405.
9. Joh JS, Lee CH, Lee JE, et al. The interval between initiation of anti-tuberculosis
treatment in patients with culture-positive pulmonary tuberculosis and receipt
of drug-susceptibility test results. J Korean Med Sci 2007;22:26e29.
10. Kim DK, Lee SW, Yoo CG, et al. Clinical characteristics and treatment responses
of tuberculosis in patients with malignancy receiving anticancer chemo-
therapy. Chest 2005;128:2218e2222.
11. Kim HJ, Kang CH, Kim YT, et al. Prognostic factors for surgical resection in
patients with multidrug-resistant tuberculosis. Eur Respir J 2006;28:576e580.
12. Kim HJ, Lee HJ, Kwon SY, et al. The prevalence of pulmonary parenchymal
tuberculosis in patients with tuberculous pleuritis. Chest 2006;129:1253e1258.
13. Kim HR, Hwang SS, Kim HJ, et al. Impact of extensive drug resistance on
treatment outcomes in non-HIV-infected patients with multidrug-resistant
tuberculosis. Clin Infect Dis 2007;45:1290e1295.
14. Lee CH, Kim WJ, Yoo CG, et al. Response to empirical anti-tuberculosis treat-
ment in patients with sputum smear-negative presumptive pulmonary
tuberculosis. Respiration 2005;72:369e374.
15. Lee HS, Oh JY, Lee JH, et al. Response of pulmonary tuberculomas to anti-
tuberculous treatment. Eur Respir J 2004;23:452e455.
16. Lee SW, Kang YA, Yoon YS, et al. The prevalence and evolution of anemia
associated with tuberculosis. J Korean Med Sci 2006;21:1028e1032.
17. Hwang SS, Kim HR, Kim HJ, et al. Impact of resistance to first-line and injectable
drugs on treatment outcomes in MDR-TB. Eur Respir J 2009;33:581e585.
18. Kim DH, Kim HJ, Park SK, et al. Treatment outcomes and long-term survival in
patients with extensively drug-resistant tuberculosis. Am J Respir Crit Care Med
2008;178:1075e1082.
19. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for
nosocomial infections, 1988. Am J Infect Control 1988;16:128e140.
20. Siegel JD, Rhinehart E, Jackson M, Chiarello L. Management of multidrug-
resistant organisms in health care settings, 2006. Am J Infect Control
2007;35:S165e193.
21. Pop-Vicas AE, D’Agata EM. The rising influx of multidrug-resistant gram-
negative bacilli into a tertiary care hospital. Clin Infect Dis 2005;40:1792e1798.
22. Ostrowsky BE, Venkataraman L, D’Agata EM, Gold HS, DeGirolami PC,
Samore MH. Vancomycin-resistant enterococci in intensive care units: high
frequency of stool carriage during a non-outbreak period. Arch Intern Med
1999;159:1467e1472.
23. Dahms RA, Johnson EM, Statz CL, Lee JT, Dunn DL, Beilman GJ. Third-generation
cephalosporins and vancomycin as risk factors for postoperative vancomycin-
resistant enterococcus infection. Arch Surg 1998;133:1343e1346.
24. Washio M, Mizoue T, Kajioka T, et al. Risk factors for methicillin-resistant
Staphylococcus aureus (MRSA) infection in a Japanese geriatric hospital. Public
Health 1997;111:187e190.
25. Wong SS, Ho PL, Yuen KY. Evolution of antibiotic resistance mechanisms and
their relevance to dialysis-related infections. Perit Dial Int 2007;27(Suppl. 2):
S272e280.