ORIGINAL ARTICLE 10.1111/j.1469-0691.2005.01257.

Stenotrophomonas maltophilia: antimicrobial resistance and molecular

typing of an emerging pathogen in a Turkish university hospital
D. Gülmez and G. Hasçelik

Department of Microbiology and Clinical Microbiology, Faculty of Medicine, Hacettepe University,

Ankara, Turkey

ABSTRACT
Despite its limited pathogenicity, Stenotrophomonas maltophilia is an emerging nosocomial pathogen. This
study investigated the isolation frequency, antimicrobial resistance and genotypic relationships of 205
S. maltophilia isolates from 188 patients in a university hospital between 1998 and 2003. Susceptibility
profiles for 11 antimicrobial agents were determined by the NCCLS agar dilution method for non-
fermentative bacteria, while enterobacterial repetitive intergenic consensus sequence (ERIC)-PCR and
pulsed-field gel electrophoresis (PFGE) were used for genotyping of the isolates. Of the 205 isolates,
56.1% were isolated in the last 2 years of the study. The risk of S. maltophilia isolation was higher in
intensive care units, S. maltophilia was isolated mostly (86.8%) after hospitalisation for ‡ 48 h, and 90.4%
of the patients had underlying diseases. Resistance levels were > 60% for all antimicrobial agents tested
except co-trimoxazole. High genetic diversity was found among the S. maltophilia isolates, and cross-
infection with S. maltophilia was not common. Although ERIC-PCR revealed fewer genotypes than
PFGE, it proved to be a rapid and easy method for S. maltophilia genotyping, and was more economical
than PFGE.
Keywords Antibiotic resistance, ERIC-PCR, genotyping, nosocomial infection, PFGE, Stenotrophomonas maltophilia
Original Submission: 20 October 2004; Revised Submission: 21 March 2005; Accepted: 17 May 2005
Clin Microbiol Infect 2005; 11: 880–886

meningitis, cholangitis, soft tissue infection and

INTRODUCTION
wound infection [3–7]. Predisposing factors for
Stenotrophomonas maltophilia is an opportunistic S. maltophilia infection include prolonged hospi-
pathogen of increasing importance. The use of talisation, especially in ICUs, consumption of
broad-spectrum antibiotics and an increase in the broad-spectrum antibiotics, malignancy, immune
number of invasive procedures and immuno- suppression, and breakdown of mucocutaneous
suppressed patients has caused this intrinsic- defence barriers (e.g., following catheterisation,
ally multidrug-resistant microorganism to artificial implants, tracheotomy, or peritoneal
emerge as an infectious agent in hospitals, dialysis) [8,9].
especially in intensive care units (ICUs) [1–3]. Epidemiological studies of clinical S. maltophilia
Its resistance to many antimicrobial agents, isolates have shown genetic diversity [10,11],
including b-lactams and aminoglycosides, allows probably associated with selection of naturally
patient colonisation even when antimicrobial present S. maltophilia from among other bacteria
agents are being used [4]. Despite its relatively by antibiotic pressure. However, cross-infections
low virulence, S. maltophilia can cause a wide between patients, transmitted by healthcare
variety of infections, e.g., pneumonia, bacter- workers, have also been reported [12]. For this
aemia, endocarditis, urinary tract infection, reason, detection of antibiotic resistance patterns
and typing of S. maltophilia isolates is significant
in the context of hospital infection control. The
Corresponding author and reprint requests: D. Gülmez, present study investigated antibiotic resistance
Hacettepe University Faculty of Medicine, Department of
Microbiology and Clinical Microbiology, Morfoloji Binasý. kat,
patterns and genotypes among S. maltophilia
06100 Sýhhiye, Ankara, Turkey isolates in a hospital for adults during the period
E-mail: [email protected] 1998–2003.

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases
 Gülmez and Hasçelik S. maltophilia resistance and typing 881

MATERIALS AND METHODS Pulsed-field gel electrophoresis (PFGE) analysis

Bacterial isolates Preparation of agarose plugs containing chromosomal DNA for

PFGE analysis was performed as described previously [18]. The
S. maltophilia isolates were obtained from various clinical DNA contained in the plugs was digested with 20 U of XbaI
specimens at the Hacettepe University Faculty of Medicine (Roche Diagnostics) at 37C overnight as recommended by the
Adult Hospital, Clinical Pathology Laboratory between 1998 manufacturer. The digested plugs and molecular size markers
and 2003. In total, 205 isolates from 188 patients were (Pulse Marker 50–1000 kb; Sigma-Aldrich) were analysed by
included in the study. Isolates from the same patient were PFGE in high gel strength agarose (Genaxis Biotechnology,
obtained from different anatomical sites. The isolates Spechbach, Germany) 1.1% w ⁄ v gels in a GN Controller
were identified initially by the Sceptor (Becton-Dickinson, (Amersham Pharmacia Biotech, Freiburg, Germany) with Tris-
Franklin Lakes, NJ, USA) system, and the identification was borate-EDTA (TBE; 0.5 M Tris, 0.5 M boric acid, 0.01 M EDTA,
confirmed by manual biochemical tests (Gram’s stain, cat- pH 8) buffer. PFGE was for 22 h at 150 V ⁄ cm at 12C, with a
alase, oxidase, aesculin hydrolysis, lysine decarboxylase and pulse time that increased from 10 s to 90 s. The gel was stained
DNase). with ethidium bromide 0.1% w ⁄ v and the DNA patterns were
evaluated using the UV Products Gel Documentation System.
Antimicrobial susceptibility testing The band patterns were interpreted according to the criteria of
Tenover et al. [19], with patterns that differed by two or three
Susceptibility to 11 antimicrobial agents (imipenem, merope-
bands being defined as closely related subtypes.
nem, co-trimoxazole, amikacin, gentamicin, ciprofloxacin,
ceftazidime, cefepime, cefotaxime, piperacillin and piperacil-
lin–tazobactam) was determined by the NCCLS agar dilution Statistics
method for non-fermentative bacteria [13]. MICs were deter- Data were analysed with SPSS software for Windows (SPSS
mined after incubation for 24 and 48 h on Mueller–Hinton Inc., Chicago, IL, USA) by the McNemar, chi-square and
agar plates at 36C. Intermediately-resistant isolates Fisher–Freeman Halton tests, with p < 0.05 being considered
were considered to be resistant. Pseudomonas aeruginosa significant.
ATCC 27853 and Escherichia coli ATCC 35218 (for piperacil-
lin–tazobactam) were included as quality control strains in
each run. RESULTS

Enterobacterial repetitive intergenic consensus sequence Isolates

(ERIC)-PCR typing
The 205 S. maltophilia isolates were from 104 male
The ERIC-PCR method used for genotyping S. maltophilia and 84 female patients. There was a gradually
isolates was optimised from previous studies [14,15]. A
single colony was inoculated into Mueller–Hinton broth and
increasing frequency of S. maltophilia isolation
incubated for 20 h at 37C. After centrifugation at 10 000 g during the study period. Of all non-fermentative
for 10 min, each pellet was washed three times in 750 lL TE bacterial isolates, S. maltophilia accounted for
buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) and then 4.0%, 5.8% and 9.7% in 2001, 2002 and 2003,
resuspended in 500 lL TE buffer. The solution was boiled
respectively, with 56.1% of the isolates being
for 20 min and centrifuged at 10 000 g for 10 min, and the
supernatant was then used as a crude DNA extract in PCRs. obtained during the last 2 years of the study. The
Amplification reactions were performed in a final volume of most frequent site of isolation was the respiratory
25 lL, with 15.5 lL H2O, 2.5 lL 25 mM MgCl2, 2.0 lL each tract (40%), followed by blood (21.5%) and pus
2.5 mM dATP, dTTP, dGTP and dCTP, 0.3 lL 1 mM ERIC2 (13.2%). S. maltophilia was the only microorgan-
primer (5¢-AAGTAAGTGACTGGGGTGAGCG-3¢) (Trilink
Biotechnologies, San Diego, CA, USA), 2.6 lL 10· PCR
ism isolated from 97 (47.3%) specimens. The other
buffer (Sigma-Aldrich, Munich, Germany), 0.13 lL Taq infections were polymicrobial. Sixty-two of the
polymerase (Sigma-Aldrich) and 2 lL DNA extract. A specimens yielded two organisms, 39 yielded
negative control with H2O instead of DNA extract was used three, and seven yielded four. The most frequent
in each run. PCRs comprised one cycle for 3 min at 94C,
co-isolated microorganisms were P. aeruginosa
two cycles of 45 s at 94C, 1 min at 30C and 1 min at 72C,
and 44 cycles of 30 s at 94C, 30 s at 55C and 1 min at (24.7%), Staphylococcus aureus (20.1%), Klebsiella
72C, with a final extension for 4 min at 72C. The PCR spp. (12.1%) and Acinetobacter spp. (10.3%).
products were analysed by electrophoresis with DNA Only 12 (6.4%) of the patients were not hospi-
Molecular Weight Marker XIV (Roche Diagnostics, Istanbul, talised, and 178 (86.8%) of the 205 isolates were
Turkey) for 2 h at 110 V in an agarose 1.5% w ⁄ v gel and
staining with ethidium bromide 0.1% w ⁄ v. The amplicon
obtained from patients after hospitalisation for
patterns were evaluated with a Gel Documentation System ‡ 2 days (Table 1). Eighty-one (46%) patients
(UV Products, Upland, CA, USA). Differences of two or were in medical wards, 43 (24.4%) in surgical
more DNA bands were considered to represent different wards, and 52 (29.6%) in ICUs. The greatest risk
strains, while isolates differing by only one band were
for S. maltophilia isolation was in ICUs if the
considered to be subtypes [11,16,17].

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 11, 880–886
882 Clinical Microbiology and Infection, Volume 11 Number 11, November 2005

Table 1. Distribution of patients and mean duration of Antimicrobial resistance

hospitalisation before isolation of Stenotrophomonas malto-
philia The resistance rates of the S. maltophilia isolates
Mean duration (days)
were > 60% for all antimicrobial agents except
of hospitalisation co-trimoxazole (Table 2). The differences between
Patients before isolation of
n (%) S. maltophilia resistance rates obtained after incubation for 24 h
and 48 h were significant for co-trimoxazole,
Medical wards 81 (46.0) 32.5
Surgical wards 43 (24.4) 39.6 ciprofloxacin, ceftazidime, cefepime, piperacillin
Intensive care units 52 (29.6) 38.3 and piperacillin–tazobactam (p < 0.05). The dif-
(medical and surgical)
Total 176 (100) 36.4
ferences in the MIC50 and MIC90 values after
incubation for 24 h and 48 h were at most two-fold,
except the MIC90 value for ciprofloxacin, which
increased four-fold. There were no significant
number of beds per unit was taken into consid- changes in the antimicrobial resistance rates dur-
eration. The mean duration of hospitalisation ing the study period, except for imipenem
before S. maltophilia was isolated was similar for (p 0.0003), meropenem (p 0.005), co-trimoxazole
all wards (Table 1). (p 0.0003), piperacillin (p 0.0001) and piperacillin–
Patient records were available for all but 13 tazobactam (p 0.0001) (Table 3).
patients. In total, 170 (90.4%) of the patients from
whom S. maltophilia was isolated had underlying
Genotyping
diseases, and 134 (71.3%) had more than one
underlying disease. Malignant diseases were the Among the 205 isolates studied, ERIC-PCR analy-
most common (35.1% of patients), followed by sis revealed 180 genotypes and PFGE analysis
hypertension (22.9%), obstructive lung disease revealed 188 genotypes. Fifteen patients yielded
(20.2%) and diabetes mellitus (17.6%). One more than one isolate, but only eight of these
patient without underlying disease had keratitis shared similar ERIC-PCR and PFGE patterns. The
caused by contamination of contact lenses. isolates with similar patterns were isolated mostly

Table 2. Resistance of Stenotropho-

24 h 48 h
monas maltophilia isolates to 11 anti-
Resistant Resistant microbial agents
Antimicrobial MIC range isolates MIC50 MIC90 isolates MIC50 MIC90
agent (mg ⁄ L) n (%) (mg ⁄ L) (mg ⁄ L) n (%) (mg ⁄ L) (mg ⁄ L)

Imipenem 0.5–1024 202 (98.5) 512 512 203 (99.0) 512 1024
Meropenem 0.5–512 201 (98.0) 128 256 201 (98.0) 128 256
Co-trimoxazole 0.25 ⁄ 4.75–32 ⁄ 608 58 (28.3) 2 ⁄ 38 8 ⁄ 152 73 (35.6) 2 ⁄ 38 8 ⁄ 152
Amikacin 2–1024 174 (84.9) 128 512 176 (85.8) 128 512
Gentamicin 2–2048 194 (94.6) 128 512 196 (95.6) 128 512
Ciprofloxacin 0.5–256 189 (92.2) 4 8 199 (97.1) 4 32
Ceftazidime 1–512 146 (71.2) 32 256 153 (74.6) 64 256
Cefotaxime 1–512 196 (95.6) 128 256 197 (96.1) 128 256
Cefepime 2–128 126 (61.5) 16 32 159 (77.6) 16 32
Piperacillin 4–2048 184 (89.8) 128 1024 197 (96.1) 256 1024
Piperacillin– 2 ⁄ 4–1024 ⁄ 4 180 (87.8) 128 512 193 (94.2) 128 512
tazobactam

Table 3. Susceptibility of Stenotrophomonas maltophilia isolates to 11 antimicrobial agents during the study period (1998–
2003)
Year IMP MER T⁄S AK GEN CIP CAZ CTX FEP PIP P⁄T
(n) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%)

1998 (16) 100.0 100.0 12.5 75.0 100.0 100.0 62.5 93.7 62.5 93.7 87.5
1999 (41) 100.0 100.0 58.5 82.9 90.2 90.2 68.3 97.6 56.1 82.9 97.6
2000 (15) 80.0 86.6 26.6 60.0 86.6 86.6 53.3 93.3 40.0 66.7 86.6
2001 (18) 100.0 100.0 16.6 88.8 94.4 94.4 72.2 83.3 55.5 50.0 50.0
2002 (42) 100.0 100.0 16.6 88.1 90.5 88.1 80.9 97.6 64.3 97.6 92.9
2003 (73) 100.0 100.0 24.6 89.0 98.6 95.9 71.2 97.3 68.5 90.4 76.7
Total 98.5 98.0 28.3 84.9 94.6 92.2 71.2 95.6 61.5 89.8 87.8

IMP, imipenem; MER, meropenem; T ⁄ S, co-trimoxazole; AK, amikacin; GEN, gentamicin; CIP, ciprofloxacin; CAZ, ceftazidime; CTX, cefotaxime; FEP, cefepime; PIP,
piperacillin; P ⁄ T, piperacillin–tazobactam.

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 11, 880–886
 Gülmez and Hasçelik S. maltophilia resistance and typing 883

Table 4. Data for 42 isolates of

ERIC-PCR PFGE Isolate Patient
Stenotrophomonas maltophilia show- pattern no. pattern no. Antibiogram no. no. Ward Date
ing their distribution into 17 ERIC-
PCR and 13 PFGE pattern groups, E1 P1 PIP, P ⁄ T different 21 H16 ICU 26 ⁄ 2 ⁄ 1999
E1 P1 26 H1 ICU 19 ⁄ 4 ⁄ 1999
antibiogram similarities, patients, E1 P1 51 H1 ICU 20 ⁄ 4 ⁄ 1999
hospital wards and isolation dates E2 P2 Similar 22 H17 ICU 12 ⁄ 3 ⁄ 1999
E2 P2 23 H18 ICU 15 ⁄ 3 ⁄ 1999
E2 P2 24 H19 ICU 15 ⁄ 3 ⁄ 1999
E3 P14 CAZ different 54 H20 ICU 25 ⁄ 11 ⁄ 1999
E3 P15 56 H21 MW 85 28 ⁄ 8 ⁄ 1999
E4a P3 Similar 64 H22 SICU 27 ⁄ 10 ⁄ 2000
E4b P3 65 H23 SW 51 16 ⁄ 11 ⁄ 2000
E5 P16 Similar 70 H24 MW 76 4 ⁄ 5 ⁄ 2000
E5 P17 71 H25 ICU 3 ⁄ 5 ⁄ 2000
E6 P18 Similar 80 H26 SW 72 28 ⁄ 2 ⁄ 2001
E6 P19 81 H27 SW 73 4 ⁄ 1 ⁄ 2001
E7 P20 Similar 87 H28 SW 53 2 ⁄ 11 ⁄ 2001
E7 P21 89 H29 SICU 6 ⁄ 12 ⁄ 2001
E8 P4 Similar 84 H30 ICU 17 ⁄ 4 ⁄ 2001
E8 P4 85 H31 ICU 29 ⁄ 5 ⁄ 2001
E9 P22 GEN, CIP, PIP, P ⁄ T different 93 H32 SW 72 9 ⁄ 2 ⁄ 2002
E9 P23 94 H33 SW 51 12 ⁄ 3 ⁄ 2002
E10a P5 Similar 102 H34 Burn 26 ⁄ 4 ⁄ 2002
E10b P5 103 H35 SW 73 1 ⁄ 5 ⁄ 2002
E11 P6 Similar 111 H2 SICU 7 ⁄ 9 ⁄ 2002
E11 P6 118 H2 SICU 12 ⁄ 10 ⁄ 2002
E11 P7 Similar 113 H3 MW 75 20 ⁄ 9 ⁄ 2002
E11 P7 116 H3 MW 75 4 ⁄ 10 ⁄ 2002
E11 P7 115 H36 SW 52 4 ⁄ 10 ⁄ 2002
E12 P8 CIP different 122 H37 ICU 16 ⁄ 11 ⁄ 2002
E12 P8 123 H38 ICU 19 ⁄ 11 ⁄ 2002
E13a P9a Similar 137 H4 SW 84 10 ⁄ 1 ⁄ 2003
E13b P9b 141 H4 SW 84 31 ⁄ 1 ⁄ 2003
E13b P9c 146 H4 SW 84 7 ⁄ 2 ⁄ 2003
E14 P10 T ⁄ S, GEN different 154 H5 MW 85 4 ⁄ 4 ⁄ 2003
E14 P10 156 H5 MW 85 7 ⁄ 4 ⁄ 2003
E15 P11 Similar 181 H6 SW 74 20 ⁄ 9 ⁄ 2003
E15 P11 190 H6 SW 74 2 ⁄ 11 ⁄ 2003
E16 P12a Similar 185 H7 SW 43 27 ⁄ 9 ⁄ 2003
E16 P12b 186 H7 SW 43 27 ⁄ 9 ⁄ 2003
E16 P12b 187 H7 SW 43 27 ⁄ 9 ⁄ 2003
E17a P13 191, 195 similar 191 H8 ICU 30 ⁄ 10 ⁄ 2003
E17a P13 194 CAZ different 195 H8 ICU 3 ⁄ 11 ⁄ 2003
E17b P24 194 H39 MW 86 3 ⁄ 11 ⁄ 2003

ICU, medical intensive care unit; SICU, surgical intensive care unit; MW, medical ward; SW, surgical ward;
PIP, piperacillin; P ⁄ T, piperacillin ⁄ tazobactam; CAZ, ceftazidime; GEN, gentamicin; CIP, ciprofloxacin; T ⁄ S,
co-trimoxazole.

from ICUs. The isolates from different wards with This organism is ubiquitous in the environment
similar ERIC-PCR patterns showed different PFGE and in the hospital setting [4,9]. Since it is able to
patterns. Antibiogram patterns were found to be grow in many different media in the presence of
unrelated to the genotypes. When MIC values most antimicrobial agents, S. maltophilia is isola-
within two dilutions were considered to be similar, ted with increasing frequency as a nosocomial
the results obtained were inconsistent with those pathogen. The annual isolation rate per 10 000
obtained by genotyping. In addition, variations in patient discharges rose from 7.1 in 1981 to 14.1 in
MICs within a genotype were observed, while 1984 at a university hospital in the USA [20]. A
isolates with different ERIC-PCR and PFGE pat- widespread study between 1997 and 2001, inclu-
terns sometimes had similar MIC values. Table 4 ding data from Asia-Pacific, Europe and America,
presents data for 42 isolates with similar ERIC-PCR showed that S. maltophilia was the third most
patterns, together with their PFGE patterns, anti- frequently isolated non-fermentative bacterium,
biogram similarities, and data concerning the following P. aeruginosa and Acinetobacter, with a
patients from whom they were isolated (including rate of isolation from clinical specimens of 8%
hospital wards and dates of isolation). [21]. As described above, the isolation frequency
of S. maltophilia increased during the period of the
present study, but further investigations are
DISCUSSION
needed to clarify the underlying reasons for this
S. maltophilia causes infections mainly in hospitals increase. As in the present study, S. maltophilia is
and is a particular risk for debilitated patients. isolated most often from respiratory specimens

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 11, 880–886
884 Clinical Microbiology and Infection, Volume 11 Number 11, November 2005

and blood. Thus, Valdezate et al. [22] described When an isolate is identified as S. maltophilia, co-
105 S. maltophilia isolates obtained between 1995 trimoxazole, ticarcillin–clavulanate, doxycycline,
and 1998, 79 of which were from the respiratory minocycline and the newer quinolones, such as
tract and 19 from blood. ofloxacin, levofloxacin, sparfloxacin and moxi-
Differentiation between S. maltophilia colonisa- floxacin, may be possible options for treatment
tion and infection may be difficult when [21,27].
S. maltophilia is not the only organism isolated. Although the NCCLS [13] suggests the use of
Sattler et al. [23] investigated episodes of infection dilution methods for testing antimicrobial sus-
from non-respiratory sites and reported that ceptibilities of S. maltophilia, the correlation
70.6% of S. maltophilia isolates were from poly- between in-vitro resistance and the clinical
microbial cultures, which yielded mostly P. aeru- response is unknown. The incubation time and
ginosa and Acinetobacter baumannii. Isolation of temperature for susceptibility testing remain con-
S. maltophilia from polymicrobial cultures may be troversial, with an increase in incubation time
related to a true infection, and is an important influencing the resistance rates of S. maltophilia for
consideration in determining initial treatment, co-trimoxazole, ciprofloxacin, b-lactams and ami-
since b-lactamases leaking from S. maltophilia cells noglycosides [28]. Garrison et al. [29] demonstra-
can facilitate the survival of b-lactam-susceptible ted that if S. maltophilia strains were incubated for
microorganisms [24]. The present study found > 24 h, mutants resistant to ticarcillin–clavulanate,
that S. maltophilia was the only microorganism ciprofloxacin and gentamicin, and which shared
isolated after cultivation of 97 (47.3%) specimens. PFGE patterns with the susceptible strains, could
The most frequent co-isolated microorganisms emerge. In the present study, the differences
from other specimens were P. aeruginosa (24.7%), between resistance rates obtained after 24 and
Staph. aureus (20.1%), Klebsiella spp. (12.1%), and 48 h of incubation were significant for co-trim-
Acinetobacter spp. (10.3%). Thus, almost half of the oxazole, ciprofloxacin, ceftazidime, cefepime, pip-
S. maltophilia isolates were monobacterial and eracillin and piperacillin–tazobactam (p < 0.05).
more likely to be a cause of infection than of As was observed in this study, S. maltophilia
colonisation. isolates have high genetic diversity, even when
The many risk-factors that predispose to isolated in a single hospital [10,11,26]. It has been
the development of S. maltophilia infection suggested that most isolates are acquired inde-
include prolonged hospitalisation, especially in pendently rather than as a consequence of cross-
ICUs, consumption of broad-spectrum antibiotics, transmission [30]. The present study showed that
malignancy, immune suppression, and a break- strains isolated from different wards and sharing
down in mucocutaneous defence barriers (e.g., the same ERIC-PCR patterns were different by
following catheterisation, artificial implants, PFGE. Although PFGE is recognised as a more
tracheostomy, or peritoneal dialysis) [2–4,8,25]. reliable method for genotyping, ERIC-PCR can
Most of the patients (90.4%) in the present study provide useful results if demographic data are
had underlying diseases, including 35.1% who also available. ERIC-PCR is a rapid and easy
had malignant diseases. These results are in method with a lower cost than PFGE.
accordance with previously published data. Cross-infections between patients are rare, but
S. maltophilia is resistant to a wide spectrum of cannot be eliminated if the patients sharing
antimicrobial agents. Berg et al. [26] investigated isolates with identical PFGE patterns are epidem-
both clinical and environmental isolates, and iologically linked [17]. The present study found
showed that the resistance profile of a strain did that only 42 isolates were genetically related
not depend on its source. In a worldwide sur- according to ERIC-PCR, and only 31 according
veillance study that included 1488 isolates to PFGE. In some cases, isolates from the same
obtained between 1997 and 2001 [21], resistance patient showed different ERIC-PCR and ⁄ or PFGE
to the antimicrobial agents tested was > 50%, with patterns. In seven of the 15 patients yielding more
the exception of co-trimoxazole (5%), gatifloxacin than one isolate from different body sites, the
(5%), levofloxacin (6%), ticarcillin–clavulanate isolates belonged to different genotypes. Isolates
(14%) and ceftazidime (34%). Similarly, the pre- belonging to the same genotype were mostly
sent study found resistance rates of > 60% for obtained from ICUs. Nosocomial outbreaks
all antimicrobial agents except co-trimoxazole. of S. maltophilia infection have been reported

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 11, 880–886
 Gülmez and Hasçelik S. maltophilia resistance and typing 885

previously. Garcia de Viedma et al. [12] typed 9. Villarino ME, Stevens LE, Schable B et al. Risk factors for
isolates from seven patients in a neonatology epidemic Xanthomonas maltophilia infection ⁄ colonisation in
intensive care unit. Infect Cont Hosp Epidemiol 1992; 13:
ward using arbitrary-primed (AP)-PCR, ERIC- 201–206.
PCR and PFGE, and were able to identify an 10. Valdezate S, Vindel A, Martin-Davila P, Del Saz BS,
index case. Similarly, Davin-Regli et al. [16] found Baquero F, Canton R. High genetic diversity among
that two patients from different wards with Stenotrophomonas maltophilia strains despite their origin-
ating at a single hospital. J Clin Microbiol 2004; 42: 693–699.
isolates which shared an AP-PCR pattern had
11. Barbier-Frebourg N, Boutiba-Boubaker I, Nouvellon M,
been in contact with the same X-ray technician. Lemeland JF. Molecular investigation of Stenotrophomonas
However, the present study of a large number of maltophilia isolates exhibiting rapid emergence of ticarcil-
isolates from a Turkish hospital found that cross- lin–clavulanate resistance. J Hosp Infect 2000; 45: 35–41.
infections with S. maltophilia were uncommon, 12. Garcia de Viedma D, Marin M, Cercenado E, Alonso R,
Rodriguez-Creixems M, Bouza E. Evidence of nosocomial
and that minor outbreaks, especially those occur- Stenotrophomonas maltophilia cross-infection in a neonatol-
ring in ICUs, can be controlled with standard ogy unit analyzed by three molecular typing methods.
precautions. Nevertheless, the frequency of isola- Infect Cont Hosp Epidemiol 1999; 20: 816–820.
tion of S. maltophilia increased during the 6-year 13. National Committee for Clinical Laboratory Standards.
Methods for dilution antimicrobial susceptibility tests for bac-
period of the study, and the management of teria that grow aerobically, 4th edn. Approved Standard M7-
infections caused by this bacterium could become A4. Villanova, PA: NCCLS, 1997.
a problem because of the multiresistant pheno- 14. Köseoğlu Ö, Sener B, Gür D. Çocuk hastalardan izole
type of these bacteria. edilen S. maltophilia suşlarının moleküler epidemiyolojisi.
Mikrobiyol Bul 2004; 38: 9–19.
15. Chatelut M, Dournes JL, Chabanon G, Marty N. Epide-
ACKNOWLEDGEMENTS miological typing of Stenotrophomonas (Xanthomonas)
maltophilia by PCR. J Clin Microbiol 1995; 33: 912–914.
This study was supported by Hacettepe University Scientific 16. Davin-Regli Bollet C, Auffray JP, Saux P, Micco D. Use of
Research Unit (Project no. 03 D 03 101004). This work was amplified polymorphic DNA for epidemiological typing of
presented, in part, at the 104th General Meeting of the Stenotrophomonas maltophilia. J Hosp Infect 1996; 32: 39–50.
American Society for Microbiology (New Orleans, LA, USA; 17. Krzewinski JW, Nguyen CD, Foster JM, Burns JL. Use of
May 2004). random amplified polymorphic DNA PCR to examine
epidemiology of Stenotrophomonas maltophilia and Achro-
mobacter (Alcaligenes) xylosoxidans from patients with cystic
REFERENCES fibrosis. J Clin Microbiol 2001; 39: 3597–3602.
18. Köseoğlu Ö, Sener B, Gülmez D, Altun B, Gür D. Steno-
1. Graff GR, Burns JL. Factors affecting the incidence of Ste-
trophomonas maltophilia as a nosocomial pathogen. New
notrophomonas maltophilia isolation in cystic fibrosis. Chest Microbiol 2004; 27: 273–279.
2001; 121: 1754–1760. 19. Tenover FC, Arbeit RD, Goering RV et al. Interpreting
2. Crispino M, Boccia MC, Bagattini M, Villari P, Triassi M, chromosomal DNA restriction patterns produced by
Zarrilli R. Molecular investigation of Stenotrophomonas pulsed-field gel electrophoresis: criteria for bacterial strain
maltophilia in a university hospital. J Hosp Infect 2002; 52: typing. J Clin Microbiol 1995; 33: 2233–2239.
88–92. 20. Morrison AJ, Hoffmann KK, Wenzel RP. Associated mor-
3. Hanes SD, Demirkan K, Tolley E et al. Risk factors for late
tality and clinical characteristics of nosocomial Pseudo-
onset nosocomial pneumonia caused by Stenotrophomonas monas maltophilia in a university hospital. J Clin Microbiol
maltophilia in critically ill trauma patients. Clin Infect Dis 1986; 24: 52–55.
2002; 35: 228–235. 21. Jones RN, Sader HS, Beach ML. Contemporary in vitro
4. Denton M, Kerr KG. Microbiological and clinical aspects of
spectrum of activity summary for antimicrobial agents
infection associated with Stenotrophomonas maltophilia. Clin tested against 18569 strains of non-fermentative Gram-
Microbiol Rev 1998; 11: 57–80. negative bacilli isolated in the SENTRY antimicrobial
5. Vidal F, Mensa J, Almela M et al. Bacteremia in adults due surveillance program. Int J Antimicrob Agents 2003; 22: 551–
to glucose non-fermentative Gram-negative bacilli other 556.
than P. aeruginosa. Q J Med 2003; 96: 227–234. 22. Valdezate S, Vindel A, Loza E, Baquero F, Canton R.
6. Çaylan R, Aydin K, Köksal I. _ Meningitis caused by Ste-
Antimicrobial susceptibilities of unique Stenotrophomonas
notrophomonas maltophilia: case report and review of the maltophilia clinical strains. Antimicrob Agents Chemother
literature. Ann Saudi Med 2002; 22: 216–218. 2001; 45: 1581–1584.
7. Aydin K, Köksal I, _ Kaygusuz S, Çaylan R, Özdemir R.
23. Sattler CA, Mason EO, Kaplan SL. Nonrespiratory Stenot-
Endocarditis caused by Stenotrophomonas maltophilia. Scand rophomonas maltophilia infection at a children’s hospital.
J Infect Dis 2000; 32: 427–430.
Clin Infect Dis 2000; 31: 1321–1330.
8. Koneman EW, Allen SD, Janda WM et al. The non- 24. Kataoka D, Fujiwara H, Kawakami T et al. The indirect
fermentative Gram-negative bacilli. In: Color atlas and pathogenity of Stenotrophomonas maltophilia. Int J Antimic-
textbook of diagnostic microbiology, 5th edn. Philadelphia: rob Agents 2003; 22: 601–606.
Lippincott-Raven, 1997; 253–320.

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 11, 880–886
886 Clinical Microbiology and Infection, Volume 11 Number 11, November 2005

25. Alfieri N, Ramotar K, Armstrong P et al. Two consecutive 28. Carroll KC, Cohen S, Nelson R et al. Comparison of
outbreaks of Stenotrophomonas maltophilia (Xanthomonas various in vitro susceptibility methods for testing
maltophilia) in an intensive care unit defined by restriction Stenotrophomonas maltophilia. Diagn Microbiol Infect Dis
fragment length polymorphism typing. Infect Cont Hosp 1998; 32: 229–235.
Epidemiol 1999; 20: 553–556. 29. Garrison MW, Anderson DE, Carroll KC et al. Stenotro-
26. Berg G, Roskot N, Smalla K. Genotypic and phenotypic phomonas maltophilia: emergence of multidrug-resistant
relationships between clinical and environmental isolates strains during therapy and in an in vitro pharmacody-
of Stenotrophomonas maltophilia. J Clin Microbiol 1999; 37: namic chamber model. Antimicrob Agents Chemother 1996;
3594–3600. 40: 2859–2864.
27. Canton R, Valdezate S, Vindel A, Del Saz BS, Maiz L, 30. VanCouwenberghe C. Evidence of nosocomial Stenotro-
Baquero F. Antimicrobial susceptibility profile of molecu- phomonas maltophilia infection in a neonatology unit ana-
lar typed cystic fibrosis Stenotrophomonas maltophilia iso- lysed by three molecular typing methods. Infect Cont Hosp
lates and differences with noncystic fibrosis isolates. Ped Epidemiol 2000; 21: 433–434.
Pulmonol 2003; 35: 99–107.

2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 11, 880–886