Local Dissemination of Multidrug-Resistant Acinetobacter baumannii Clones in a Thai Hospital

Abstract

A total of 83 Acinetobacter baumannii isolates from patients attending a tertiary care university hospital in Thailand were investigated for their clonal relatedness, antimicrobial susceptibility profiles, and integron carriage. Susceptibility profiles showed that 56 (67%) of these isolates exhibited multiple drug resistance (MDR). Pulsed-field gel electrophoresis (PFGE) showed that 73% of these resistant isolates were clustered into three predominant PFGE types: 6, 7, and 36. This suggested that the high number of isolates exhibiting MDR phenotypes observed in the hospital is, to some extent, due to the spread of these three resistant clones. Class 1 integrase genes were detected in all MDR isolates belonging to PFGE type 6, most MDR isolates belonging to PFGE type 7 and none of the isolates belonging to PFGE type 36. Five different class 1 gene cassette arrays, dfrA1-orfC, bla_IMP-14-aac6′, aacA4- catB8-aadA1, aacC1-orfX-orfX′-aadA1a, and aacC1-orfX-orfX-orfX′-aadA1a, were identified, of which the bla_IMP-14-aac6′ array has only been found in Thai isolates. Two isolates identified in this study carried a class 2 integrase gene with a 2.2 kb cassette array containing aadA1-sat-dfrA1.

Introduction

A cinetobacter baumannii is one of the major causative agents of nosocomial infections in healthcare facilities worldwide.^4,12,53 This organism, like Pseudomonas aeruginosa, is intrinsically resistant to a number of antimicrobial agents and the emergence of multidrug-resistant (MDR) clinical isolates has been widely reported from hospitals in Europe, North America, Argentina, Brazil, China, Taiwan, Hong Kong, Japan, and Korea.³⁵ Multidrug resistance in A. baumannii has often been linked to the presence of integrons; the genetic elements that function as a gene capture and expression system.^{13,16,17,22,24} Integrons consist of two conserved segments (CS) at the 5′- and 3′-ends, separated by an inserted gene cassette, which may contain antimicrobial resistance genes. The 5′-CS contains an integrase gene followed by a recombination site, attI, where gene cassettes are inserted or excised by a site-specific recombination mechanism.⁴¹ Integrons are classified based on the similarity of their nucleotide sequences to those of the highly conserved class-specific integrase genes. Among these, integrons of class 1 through 5 are generally associated with mobile DNA elements such as insertion sequences, transposons and conjugative plasmids that could facilitate the transfer of genetic materials within and between bacterial populations.²⁷ At present, class 1, 2, and 3 integrons have been shown to be involved in conferring antimicrobial resistance.²⁷

In Thailand, the incidence of MDR A. baumannii has steadily increased in recent years.^2,23,36,43 Previous studies regarding A. baumannii infection in Thailand have primarily focused on the in vitro susceptibilities of clinical isolates to various antimicrobial agents as well as treatment and risk factors for infection.^6,19,23,44 Thus far, there has been only one study involving genotyping of 13 A. baumannii isolates with high-level imipenem (IPM) resistance from a regional hospital in the north of Thailand.³¹ Hence, this study aims, for the first time, at investigating integron carriage, antimicrobial susceptibility profiles and genetic relatedness among a collection of 83 Thai clinical isolates of A. baumannii from a tertiary care teaching hospital in Bangkok.

Materials and Methods

Bacterial strains and patients' data

Clinical isolates of A. baumannii from 78 patients who were treated at Ramathibodi hospital, a 1,350-bed tertiary care teaching hospital in Bangkok, Thailand, from October 2005 to January 2006 were randomly chosen. Relevant information about the patients, including age, sex, hospital number, admission department/ward, and source of specimen, were obtained by reviewing the culture request forms.

Bacterial identification

Bacteria were identified at the Division of Microbiology, Department of Pathology, Ramathibodi Hospital, by routine biochemical tests as described in the Manual of Clinical Microbiology 8th edition.²⁹ The biochemical tests included triple sugar iron agar, indole, citrate, nitrate reduction, as well as fermentation of dextrose, lactose, mannitol, maltose, sucrose, and xylose. A. baumannii was differentiated from A. calcoaceticus by growth characteristics at 42°C. Species identification was confirmed by detection of the bla_OXA51-like gene as recommended by Turton et al.⁴⁷ Each isolate was subcultured twice on trypticase soy agar (Becton Dickinson) before storage at −70°C in 2 ml of 20% skimmed milk containing 30% glycerol.

Pulsed-field gel electrophoresis and analysis of pulsed-field gel electrophoresis profiles

Genomic DNA in agarose gel plugs was prepared according to the protocol recommended by Antibiotic Resistance Prevention and Control (http://www.hpa.org.uk/web/HPAwebFile/HPAweb_C/1194947313339). In-gel digestion of genomic DNA with 20 units of ApaI (New England Biolabs) was carried out at 30°C overnight. Electrophoresis was performed on a 1% SeaKem LE agarose (FMC) gel using a CHEF-DRIII system (Bio-Rad Laboratories) at 200 V for 20 h at 14°C, with a field angle of 120°, switch times ranging from 5 to 20 s and linear ramping. For comparison of pulsed-field gel electrophoresis (PFGE) patterns, the band sizes were manually assigned after normalization with the PFG λ ladder. Cluster analysis was then performed using Dice coefficient of similarity at a tolerance setting of 3% band tolerance, and a dendrogram was generated using unweighted pair group matching band average with GeneDirectory Software (SynGene). The cutoff levels of 80% and 96% similarity were used to define PFGE types and subtypes. These cutoff levels also correspond to Tenover's criteria for visual comparison for genetic relatedness.⁴² Isolates with a similarity index of <80% were considered as unrelated and assigned to different PFGE types using numerical designations. Members of each PFGE type that have <96% similarity were further differentiated into subtypes that were designated using letters a–e.

Antimicrobial susceptibility testing

The standard agar dilution method recommended by Clinical Laboratory Standards Institute⁸ was used to determine the MICs of amikacin (AMK), ceftazidime (CAZ), ciprofloxacin (CIP), and colistin (COL), whereas Etest strips (AB Biodisk) were used to determine the MICs of IPM, cefoperazone-sulbactam (SCFP), and tigecycline (TGC). The MIC interpretive standards used were the Clinical Laboratory Standards Institute– recommended breakpoints for other nonenterobacteriaceae.⁸ For TGC, the MIC interpretive breakpoints for Enterobacteriaceae recommended by the U.S. FDA (S ≤2 μg/ml; I = 4 μg/ml; R ≥8 μg/ml) were used.³⁹

Detection of class 1, class 2, and class 3 integrase genes

The IntIF and IntIR primer pair²² was used in the PCR amplification of the class 1 integrase gene, whereas the Int2.F and Int2.R and the Int3.F and Int3.R primer pairs²⁸ were used to amplify the class 2 and class 3 integrase genes, respectively. The amplification reaction mix contained 200 ng of each primer, 200 nM dNTP, 1 U of Taq DNA polymerase (New England Biolabs), and 2.5 μl of template DNA, which was prepared by a simple heat lysis method.¹¹ The amplification reaction was carried out using an initial denaturation at 94°C for 5 min, 30 cycles of denaturation at 94°C for 1 min, annealing at 59°C for 1 min, and extension at 72°C for 1 min, and a final extension at 72°C for 7 min.¹¹

Detection and characterization of class1 and class 2 gene cassette arrays

Amplification of class 1- and class 2-inserted gene cassette arrays was performed by using the 5′-CS and 3′-CS primer pairs²⁶ and the hep74 and hep51 primer pairs,⁵¹ respectively. The amplification reactions were carried out as described above with the exception that the annealing temperature used was 55°C. For DNA sequencing of the gene cassette arrays, the major amplified products were cloned into pDrive “UA” cloning vector (Qiagen) and the purified plasmids were sent to Macrogen Inc. for sequencing. DNA sequence analysis was performed using the BLAST program from the National Centre for Biotechnology Information. Multiple sequence alignments and restriction map analyses were performed using the BioEdit software (Ibis Therapeutics).

Definition

MDR isolates were defined as those exhibiting resistance to at least three of the following five drug classes: cephalosporins (CAZ), carbapenems (IPM), fluoroquinolones (CIP), aminoglycosides (AMK), and the β-lactam-β-lactamase inhibitor combination, SCFP.³⁴

Sporadic PFGE types were defined as those with ≤3 members.

Clone was defined as a group of isolates exhibiting ≥80% similarity in their PFGE profiles and was assigned to the same PFGE type.

Results

Clinical strains and patients' data

A total of 83 isolates of A. baumannii from clinical samples of 78 patients were included in this study, which consisted of 54 males and 24 females. The age of the patients were distributed from 1 to 99 years with the majority of the patients (64%) over 50 years of age. Almost all of the patients were inpatients with 46, 16, 5, 4, 2, and 2 patients admitted to the Departments of: Medicine, Surgery, Pediatrics, Burns, Orthopedics, and Otolaryngology, respectively. Only 3 outpatients were included. Among these, two were admitted to the observation ward (Aci18 and Aci84), whereas the other one was admitted to the Department of emergency (Aci94). Of 83 isolates, 25 isolates were recovered from patients during their admission to the intensive care units (ICUs): 3IC, 4IK, 5IC, 9CC, and 9IC. Most samples (61%) were isolated from the respiratory tract including samples of sputum (42%), tracheal suction (12%), broncho-alveolar lavage (2%), lung aspirate (1%), chest drain (1%), pleural fluid (1%), and throat swab (1%). Nineteen samples (23%) were isolated from cutaneous wounds. Six isolates (Aci16, 31, 45, 71, 91, and 96) were recovered from blood cultures. Three samples were isolated from body fluids, including bile (2%) and ascitic fluid (1%). There were five pairs of isolates that were each recovered from a single patient that were included in this study. Aci52 and Aci1 were isolated from two different samples (sputum and wound, respectively) of the same patient within a gap of 20 days, whereas Aci12 and Aci13 were isolated from repeated wound samples of the same patient. Aci2 and Aci32 were isolated 24 days apart from the sputum and a tracheal brushing of an 18-year-old male, respectively. Aci8 and Aci20 were collected a week apart from wound samples of a 58-year-old female. Aci42 and Aci40 were isolated 3 days apart from repeated cultures of bile from an autopsy.

Clonal spreading of isolates belonging to PFGE types 6, 7, and 36

To determine the clonal relatedness of the isolates, PFGE typing of all 83 isolates was carried out using the cutoff levels of 80% and 96% to define PFGE types and subtypes, respectively. The typing results revealed 36 different PFGE types, 1–36, of which 30 isolates showed unique PFGE fingerprints and 7 isolates belonged to either PFGE types 1, 16, or 26 that have 2–3 members (Fig. 1). Interestingly, the remaining 46 isolates fell within three predominant PFGE types: 6, 7, and 36 (Fig. 1, isolates in boxes). The distribution of members of the predominant PFGE types in various departments of the hospital revealed both inter- and intradepartmental spread over the 3-month study period (Fig. 1, isolates in boxes). For example, PFGE type 7 consisted of 15 isolates, 12 of which were disseminated among the Department of Medicine, especially within the 7NW ward and the ICU, 9IC (Fig. 1, middle box). Departmental distribution of 13 isolates from PFGE type 6 showed that 12 isolates were disseminated among the Department of Medicine, the Burn Unit, and the Department of Surgery (Fig. 1, upper box). Likewise, out of 18 isolates belonging to PFGE type 36, 12 and 5 members were recovered from the Departments of Medicine and Surgery, respectively (Fig. 1, bottom box). In addition, the results from PFGE typing of two pairs of isolates recovered from repeated sample cultures from two patients suggested the possibility of infections with mixed bacterial populations. Aci12, belonging to PFGE subtype 7e (Fig. 1, middle box), and Aci13, belonging to PFGE subtype 36b (Fig. 1, bottom box), were recovered from repeated wound samples that were sent to the lab separately on the same day. Likewise, Aci40 belonging to PFGE subtype 36e (Fig. 1, bottom box) and Aci42 belonging to PFGE subtype 6c (Fig. 1, upper box) were isolated from bile samples from the same autopsy that were sent to the lab 1 day apart.

FIG. 1.

A dendrogram obtained from cluster analysis of the pulsed-field gel electrophoresis (PFGE) fingerprints of 83 isolates of Acinetobacter baumannii. The analysis was performed using Dice coefficient of similarity at a tolerance setting of 3% band tolerance, and a dendrogram was generated using unweighted pair group matching band average with GeneDirectory Software (SynGene). The cutoff levels of 80% and 96% similarity were used to define PFGE types and subtypes. M, Department of Medicine; P, Department of Pediatrics; S, Department of Surgery; B, The Burn Unit; OW, Observation Ward; Or, Department of Orthopedics; Aut, Autopsy. Underlined isolates indicate those exhibiting a multiple drug resistance phenotype.

Antimicrobial susceptibility profiles

Antibiogram typing of the 83 isolates performed using representatives of five antimicrobial drug classes commonly used in the treatment of Gram-negative nosocomial infections, which are AMK, CAZ, IPM, CIP, and SCFP, generated 21 antibiogram types (Table 1). Antibiogram type 1 (susceptible to all five agents) consisted of 11 isolates with unique PFGE types, whereas the remaining 3 isolates (Aci51, 52, and 81) were the members of predominant PFGE types 7 and 36 (Table 1). Similarly, antibiogram types 2–5 (resistant to 1 agent) and antibiogram types 6–9 (resistant to 2 agent) were comprised a total of 11 isolates from sporadic PFGE types, whereas only 1 isolate each was from the predominant PFGE types 36 and 7. On the other hand, antibiogram types 10–21 (MDR) consisted of 56 isolates, of which 41 isolates belonged to the three predominant PFGE types, whereas only 10 isolates had unique PFGE types. The remaining 5 MDR isolates belonged to PFGE types that have three and two members (PFGE types 16 and 26). These observations suggested that the MDR isolates were more likely to be selected and disseminated among patients. All of the 56 MDR isolates remained susceptible to COL and the two colistin resistant isolates (Aci2 and 14) found in this study belonged to antibiogram type 1 (Table 1). Only two MDR isolates (Aci57 and 53) showed intermediate resistance to TGC (data not shown) and one MDR isolate (Aci82) exhibited resistance to TGC (Table 1). Thus, our in vitro susceptibility results were in accordance with those of other studies^33,39,44,45 in that COL and TGC were effective against the majority of clinical isolates of A. baumannii, including those exhibiting multiple drug resistance. Moreover, among the 25 isolates from the ICUs, 21 isolates (∼84%) were MDR (Table 1, antibiogram types 10–21, underlined isolates).

Table 1.

Antibiogram Typing and Integron Carriage of 83 Acinetobater baumannii Isolates

	Antimicrobial agents					No. of isolates
Antibiogram type	AMK	CAZ	IPM	CIP	SCFP	Total	Integrase gene	PFGE types (strain no.)
1	S	S	S	S	S	14	1	4 (Aci16^*), 5 (Aci7), 7 (Aci51), 11 (Aci93), 14 (Aci79), 17 (Aci6), 18 (Aci94), 21 (Aci18), 22 (Aci85), 23 (Aci24), 32 (Aci14^COL), 35 (Aci2^COL), 36 (Aci52, 81)
Total no. of isolates susceptible to 5 agents						14	1
2	S	R	S	S	S	1	1	10 (Aci45^*)
3	S	S	R	S	S	1	-	36 (Aci69)
4	S	S	S	R	S	3	-	1 (Aci31), 2 (Aci65), 13 (Aci25^TGC)
5	S	I	I	R	S	1	-	1 (Aci74)
Total no. of isolates resistant to 1 agent						6	1
6	S	S	R	R	S	1	-	15 (Aci50)
7	S	R	S	R	S	4	4	7 (Aci76^), 20 (Aci55^), 30 (Aci95^), 33 (Aci72^)
8	S	R	S	R	I	1	1	28 (Aci33^*)
9	I	R	S	R	S	1	1	25 (Aci9^*)
Total no. of isolates resistant to 2 agents						7	6
10	R	R	S	R	S	1	1	16 (Aci83^*)
11	S	R	R	R	S	12	4	6 (Aci3^, 20^1+2), 7 (Aci1^, 78), 9 (Aci61^), 36 (Aci23, 37, 40, 48, 68, 87, 89)
12	I	R	R	R	S	3	1	6 (Aci8^*), 34 (Aci36), 36 (Aci64),
13	S	R	R	R	I	10	10	6 (Aci4^, 21^, 39^, 56^, 57^, 63^, 84^), 16 (Aci49^, 44^), 19 (Aci19^2)
14	I	R	R	R	I	2	1	29 (Aci22^*), 36 (Aci97)
15	R	S	R	R	I	2	-	36 (Aci32, 98)
16	R	I	R	R	I	1	-	36 (Aci13)
17	S	R	R	R	R	4	4	6 (Aci42^), 24 (Aci46^), 26 (Aci29^, 59^)
18	I	R	R	R	R	2	1	6 (Aci96^*), 36 (Aci54)
19	R	R	R	R	S	7	6	6 (Aci62^), 7 (Aci60^, 86^, 90^, 91^),27 (Aci80^), 36 (Aci53)
20	R	R	R	R	I	11	9	3 (Aci82^* ^TGC), 7 (Aci10^, 12^, 38^, 58^,73^,88^),8 (Aci5), 12 (Aci77^), 31 (Aci17^), 36 (Aci67)
21	R	R	R	R	R	1	1	7 (Aci35^*)
Total no. of multidrug-resistant isolates						56	38
Overall no.						83	46

*, *² and *¹⁺² indicate isolates with class 1, class 2, and class 1 and 2 integrase genes, respectively. Underlined isolates indicate those from the intensive care unit. ^COL and ^TGC indicate isolates that are resistant to colistin and TGC, respectively. Predominant PFGE types are in bold type.

AMK, amikacin; CAZ, ceftazidime; CIP, ciprofloxacin; IPM, imipenem; SCFP, cefoperazone-sulbactam; PFGE, pulsed-field gel electrophoresis; COL, colistin; TGC, tigecycline; S, susceptible; I, intermediate resistant; R, resistant.

Comparison of the susceptibility results suggested that isolates belonging to the predominant PFGE types exhibited higher resistance to CAZ, IPM, and CIP than those of sporadic isolates. The percentages of susceptible isolates to CAZ, IPM, and CIP were 11, 9, and 9, respectively, for the predominant PFGE types and 41, 57, and 32, respectively, for sporadic PFGE types (Table 2). All 13 isolates belonging to PFGE type 6 clone exhibited resistance to CAZ and IPM, as well as CIP with the MIC₅₀ values of 128, >32, and 64 μg/ml, respectively (Tables 2 and 3). Thirteen out of 15 members of PFGE type 7 clone were MDR, 11 of which exhibited resistance to AMK, CAZ, IPM, and CIP. The MIC₅₀ of AMK, CAZ, IPM and CIP of isolates in PFGE type 7 were >512, 64, >32, and 32 μg/ml, respectively (Tables 2 and 3). Fifteen out of 18 isolates from PFGE type 36 were MDR (Table 3). Of these 15 MDR isolates, 3 were resistant to AMK, CIP, and IPM, whereas 10 isolates showed resistance to CAZ, CIP, and IPM. The remaining 2 isolates exhibited resistance to AMK, CAZ, CIP, and IPM (Table 3). The MIC₅₀ values for AMK, CAZ, IPM, and CIP of isolates in PFGE type 36 were 16, >512, 32, and 32 μg/ml, respectively (Table 2).

Table 2.

Comparison of Antimicrobial Susceptibility for Isolates Belonging to Predominant and Sporadic Pulsed-Field Gel Electrophoresis Types

		AMK		CAZ		IPM		CIP		SCFP		TGC		COL
PFGE type	Total isolates	No. of susceptible isolates	MIC₅₀/MIC₉₀	No. of susceptible isolates	MIC₅₀/MIC₉₀	No. of susceptible isolates	MIC₅₀/MIC₉₀	No. of susceptible isolates	MIC₅₀/MIC₉₀	No. of susceptible isolates	MIC₅₀/MIC₉₀	No. of susceptible isolates	MIC₅₀/MIC₉₀	No. of susceptible isolates	MIC₅₀/MIC₉₀
Sporadic	37	28	8/256	15	64/> 512	21	1.5/> 32	12	32/128	25	8/32	33	0.75/2	35	2/2
Predominant
6	13	10	16/32	0	128/512	0	>32/>32	0	64/64	4	24/32	12	2/2	13	2/2
7	15	4	>512/>512	1	64/512	2	>32/>32	1	32/64	8	16/32	14	0.5/2	15	2/2
36	18	10	16/256	4	>512/>512	2	32/>32	3	32/128	13	12/32	17	1.5/2	18	2/2
Total	46	24	16/>512	5	128/>512	4	>32/>32	4	32/64	25	16/32	43	1.5/2	46	2/2

MIC in μg/ml.

Table 3.

Patient's Data and Characteristics of the 56 Multidrug-Resistant and Selected Non-Multidrug-Resistant Isolates

				Patient's data
PFGE type	Isolate no.	Date of isolation	Source	Age/sex	Dept. (ward)	Antibiogram type	Resistance profile	Class of integrase gene	Gene array size (kb)	Cassette array
3	Aci82	26/12/05	Sputum	88/M	Oto (6NWE)	20	AMK,CAZ, CIP, IPM, TGC	1	—
4	Aci16^a	20/11/05	Blood	40/M	M (9SW)	1	—	1	1.5	bla_IMP-14-aac(6′)
6a	Aci8	19/11/05	Wound	58/F	B (5NB)	12	CAZ, CIP, IPM	1	—
6b	Aci20	26/11/05	Wound	58/F	B (5NB)	11	CAZ, CIP, IPM	1 & 2	2.2^b	aadA1-sat-dfrA1
	Aci3	17/11/05	Pus	34/M	B (5NB)	11	CAZ, CIP, IPM	1	—
	Aci62	01/11/05	Urine	75/M	M (9SE)	19	AMK, CAZ, CIP, IPM	1	—
6c	Aci57	28/10/05	Sputum	91/M	M (7NW)	13	CAZ, CIP, IPM	1	—
	Aci63	07/11/05	Sputum	55/M	S (3IC)	13	CAZ, CIP, IPM	1	—
	Aci4	13/11/05	Sputum	85/M	M (9CC)	13	CAZ, CIP, IPM	1	—
	Aci21	26/11/05	Wound	59/M	M (9NK2)	13	CAZ, CIP, IPM	1	—
	Aci39	09/12/05	Trachea	65/M	S (5IC)	13	CAZ, CIP, IPM	1	—
	Aci42	11/12/05	Bile	56/M	M (Aut)	17	CAZ, CIP, IPM, SCFP	1	—
	Aci84	26/12/05	Sputum	6/F	OW (1OW)	13	CAZ, CIP, IPM	1	—
	Aci56	27/10/05	Lung	56/M	M (9IC)	13	CAZ, CIP, IPM	1	—
	Aci96	02/01/06	Blood	70/M	M (9IC)	18	CAZ, CIP, IPM, SCFP	1	—
7a	Aci73	18/12/05	Sputum	59/M	M (7NW)	20	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
7b	Aci60	29/10/05	Sputum	29/M	M (9IC)	19	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
7c	Aci51^a	24/10/05	Trachea	7/M	P (8NC)	1	—	—	—
	Aci76^a	18/12/05	Trachea	36/M	S (5IC)	7	CAZ, CIP	1	—
	Aci78	23/12/05	Sputum	66/M	M (9IC)	11	CAZ, CIP, IPM	—	—
	Aci58	27/10/05	Sputum	74/M	M (7NW)	20	AMK, CAZ, CIP, IPM	1	—
	Aci10	18/11/05	Sputum	81/F	M (7NW)	20	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
	Aci38	08/12/05	Sputum	74/F	M (7NW)	20	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
	Aci88	30/12/05	Urine	80/M	M (7NW)	20	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
	Aci90	31/12/05	Sputum	48/F	M (7NW)	19	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
	Aci91	29/12/05	Blood	37/F	M (9IC)	19	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
	Aci86	27/12/05	Sputum	30/M	M (4IK)	19	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
7d	Aci1	12/11/05	Wound	75/M	M (9IC)	11	CAZ, CIP, IPM	1	—
7e	Aci12	19/11/05	Wound	67/F	M (8NK2)	20	AMK, CAZ, CIP, IPM	1	3	aacC1-orfX-orfX-orfX′-aadA1a
	Aci35	08/12/05	Back	49/M	B (5NB)	21	AMK, CAZ, CIP, IPM, SCFP	1	2.4	aacA4-catB8-aadA1
8	Aci5	16/11/05	Sputum	69/M	M (4IK)	20	AMK, CAZ, CIP, IPM	—	—
9	Aci61	31/10/05	Sputum	34/M	M (9IC)	11	CAZ, CIP, IPM	1	—
10	Aci45^a	01/12/05	Blood	7/F	P (8NW)	2	CAZ	1	1.2	dfrA1-orfC
12	Aci77	22/12/05	Throat	85/F	M (9SW)	20	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
16a	Aci49	06/11/05	Urine	67/M	M (7NE)	13	CAZ, CIP, IPM	1	—
16b	Aci44	14/12/05	Sputum	71/F	M (9IC)	13	CAZ, CIP, IPM	1	—
	Aci83	24/12/05	Trachea	38/M	M (7NW)	10	AMK, CAZ, CIP	1	—
19	Aci19	23/11/05	Arm	77/M	S (5NW)	13	CAZ, CIP, IPM	2	2.2^b	aadA1-sat-dfrA1
24	Aci46	01/12/05	Pus	76/M	M (7NW)	17	CAZ, CIP, IPM, SCFP	1	—
25	Aci9^a	20/11/05	Sputum	42/M	M (7NW)	9	CAZ, CIP	1	1.2	dfrA1-orfC
26	Aci59	28/10/05	Sputum	85/F	S (5IC)	17	CAZ, CIP, IPM, SCFP	1	—
	Aci29	12/01/06	Sputum	50/F	S (7SE)	17	CAZ, CIP, IPM, SCFP	1	—
27	Aci80	22/12/05	Sputum	45/F	M (9IC)	19	AMK, CAZ, CIP, IPM	1	2.4	aacA4-catB8-aadA1
28	Aci33^a	08/12/05	Tissue	64/M	S (5NW)	8	CAZ, CIP	1	2.5	aacC1-orfX-orfX′-aadA1a
29	Aci22	28/11/05	Leg	30/M	B (5NB)	14	CAZ, CIP, IPM	1	1.2	dfrA1-orfC
31	Aci17	11/01/06	Pleural	87/M	M (7NW)	20	AMK, CAZ, CIP, IPM	1	3	aacC1-orfX-orfX-orfX′-aadA1a
34	Aci36	09/12/05	Wound	83/F	M (7NK)	12	CAZ, CIP, IPM	—	—
36a	Aci64	02/11/05	Ascitic	66/M	S (5SW)	12	CAZ, CIP, IPM	—	—
	Aci97	01/01/06	Urine	87/F	M (7NW)	14	CAZ, CIP, IPM	—	—
36b	Aci52^a	24/10/05	Sputum	75/M	M (7NW)	1	—	—	—
	Aci37	08/12/05	Sputum	82/F	S (5NW)	11	CAZ, CIP, IPM	—	—
	Aci32	07/12/05	Trachea	18/M	S (3IC)	15	AMK, CIP, IPM	—	—
	Aci13	19/11/05	Wound	67/F	M (8NK2)	16	AMK, CIP, IPM	—	—
	Aci53	25/10/05	Sputum	39/M	M (9IC)	19	AMK, CAZ, CIP, IPM	—	—
36c	Aci81^a	22/12/05	Wound	66/M	Or (2TC)	1	—	—	—
	Aci69^a	13/11/05	Sputum	76/M	M (8NK1)	3	IPM	—	—
	Aci68	12/11/05	BAL	74/M	M (7NW)	11	CAZ, CIP, IPM	—	—
	Aci89	30/12/05	Sputum	86/F	M (7NW)	11	CAZ, CIP, IPM	—	—
	Aci48	12/01/06	Sputum	47/F	M (7NW)	11	CAZ, CIP, IPM	—	—
	Aci23	02/12/05	Sputum	94/M	M (9CC)	11	CAZ, CIP, IPM	—	—
	Aci87	27/12/05	Sputum	60/M	M (9CC)	11	CAZ, CIP, IPM	—	—
36d	Aci54	27/10/05	Trachea	68/M	S (5IC)	18	CAZ, CIP, IPM, SCFP	—	—
36e	Aci40	09/12/05	Bile	56/M	M (Aut)	11	CAZ, CIP, IPM	—	—
	Aci98	03/01/06	Sputum	83/M	M (4IK)	15	AMK, CIP, IPM	—	—
	Aci67	09/11/05	Sputum	50/M	S (5SW)	20	AMK, CAZ, CIP, IPM	—	—

Non-multidrug-resistant isolates.

Class 2 gene cassette array.

M, Medicine; S, Surgery; Or, Orthopedics; OW, observation ward; P, Pediatrics; B, burn unit; Oto, Otolaryngology; BAL, broncho-alveolar lavage.

Integrase gene distribution and characterization of class 1- and class 2-inserted gene cassette arrays

In A. baumannii, it was shown that strains carrying integrons were significantly associated with multidrug resistance.^14,22 Thus, the distribution of class 1, class 2, and class 3 integrase genes was investigated using multiplex PCR amplification.¹¹ The sizes of the amplified products for class 1, class 2, and class 3 integrase genes are 160, 789, and 979 bps, respectively (data not shown). The results showed that class 1 integrase genes were found in 2 out of 20 isolates belonging to antibiogram types 1–5 (resistant to ≤1 agent), and in 6 out of 7 isolates belonging to antibiogram types 6–9 (resistant to 2 agents) (Table 1, isolates with asterisks). About 83% (38/46) of the integrase gene-carrying isolates were MDR (Table 1, antibiogram types 10–21). Among these 38 isolates, 13 isolates each were from PFGE types 6 and 7 clones (Table 3). None of the isolates belonging to PFGE type 36 carried a class 1 integrase gene. One MDR isolate (Aci19) belonging to the unique PFGE type 19 carried a class 2 integrase gene, whereas another MDR isolate (Aci20) of the predominant PFGE type 6 carried both class 1 and class 2 integrase genes. A class 3 integrase gene was not detected in any isolates.

Integron cassette PCR detected gene cassette arrays of approximately 1.2, 1.5, 2.4, 2.5, or 3 kb in only 18 out of 45 class 1 integrase gene-positive isolates (Table 3). The failure to amplify the cassette array in some class 1 integrase gene-containing isolates, including all members of PFGE type 6 clone (Table 3), could either be due to the absence of the cassette array itself or the inability of the primer set to amplify some cassette arrays.⁴⁶

A gene cassette array of approximately 2.2 kb was detected in both class 2 integrase gene positive isolates (Table 3). HaeIII and HinfI digestion of the 2.2-kb class 2 gene cassette arrays from these two isolates yielded similar restriction banding patterns (data not shown), suggesting that the gene cassettes were identical. Sequence analysis of the cassette array from Aci20 (GenBank accession no. HM055363) revealed that the 2.2-kb class 2 gene cassette array contained dfrA1, encoding dihydrofolate reductase (confers resistance to trimethoprim), sat encoding streptothicin acetyltransferase (confers resistance to streptothicin), aadA1 (confers resistance to streptomycin and spectinomycin), and orfX (ybeA).

Sequence analysis of the 1.2-kb amplicons obtained from Aci9 (GenBank accession no. HM036078), Aci22, and Aci45 showed that the 1.2 kb class 1 gene cassette arrays contained dfrA1 encoding dihydofolate reductase that confers resistance to trimethoprim and orfC, encoding a hypothetical protein of unknown function (Table 3). All three isolates belonged to unique PFGE types and only Aci22 (PFGE type 29) exhibited an MDR phenotype (Table 3).

Sequence analysis of the 1.5-kb amplicon obtained from Aci16, belonging to PFGE type 4 (GenBank accession no. HM036079), showed that the 1.5-kb class 1 gene cassette array contained bla_IMP-14, encoding the variant metallo-β-lactamase (MBL) IMP-14 and aac(6′), encoding aminoglycoside 6′-N-acetyltransferase. Repeated Etest results revealed that Aci16 was susceptible to IPM (Table 3, PFGE type 4) with an MIC of 0.38 μg/ml (data not shown).

Nine out of 13 MDR isolates belonging to PFGE type 7 and 2 MDR isolates belonging to the unique PFGE types 12 and 27 carried a class 1 integron containing a 2.4-kb gene cassette array (Table 3) that showed similar restriction banding patterns upon digestion with HaeIII and HinfI (data not shown) suggesting that the gene cassettes were identical. Sequence analysis of the amplicon from a representative isolate, Aci77 (PFGE type 12), revealed that the 2.4-kb class 1 gene cassette array contained aacA4 encoding aminoglycoside 6′-N-acetyltransferase (confers resistance to AMK, netilmycin and tobramycin), catB8 encoding chloramphenicol acetyltransferase, and aadA1 encoding aminoglycoside 3′-N-adenyltransferase (confers resistance to spectinomycin and streptomycin).

The 2.5 kb gene cassette was found in only 1 isolate (Aci33) belonging to unique PFGE type 28 (Table 3). Sequence analysis revealed that the 2.5-kb class 1 gene cassette array contained aacC1 encoding 3-N-aminoglycoside acetyltransferase (confers resistance to gentamicin), two openreading frames, orfX and orfX′ and aadA1a (GenBank accession no. HMO036080).

The 3-kb gene cassette was obtained from two MDR isolates belonging to PFGE type 7(Aci12) and PFGE type 31 (Aci17) (Table 3). Digestion of the 3-kb amplicons with HaeIII and HinfI yielded similar restriction banding patterns (data not shown), suggesting that the gene cassettes were identical. Sequence analysis of the amplicon obtained from a representative isolate, Aci12, revealed that the gene cassettes of the 3-kb array were identical to those of the 2.5-kb class 1 gene cassettes with the exception of an extra copy of orfX inserted between the orfX and orfX′ region.

Discussion

Analysis of individual strains for clonal relatedness using PFGE typing showed that 41 of 56 MDR isolates (∼73%) were clustered into three predominant PFGE types: 6, 7, and 36 (Fig. 1, underlined isolates in boxes). This suggests the spread of these three resistant clones in the hospital, especially in the Departments of Medicine and Surgery, and the Burn Unit (Fig. 1). By contrast, most of the susceptible isolates and isolates with resistance to only 1–2 of the antimicrobial agents tested were more heterogeneous in their PFGE patterns (Table 1). Hence, our results are in agreement with earlier reports^9,10,48 that outbreak strains tend to cluster and exhibit MDR phenotypes, whereas nonoutbreak strains are usually diverse and more susceptible to antimicrobial agents. Our PFGE typing results also revealed the existence of subtypes among the predominant PFGE types. This suggests the possibility of prolonged persistence of the original strain over the course of time causing gradual alterations in its genetic content. Moreover, variations in the susceptibility profiles were observed among members of PFGE types 7c, 36b, and 36c, which consisted of both susceptible and multiple drug-resistant populations (Table 3). It appeared that the resistant strains might have acquired additional resistance determinants that favor their survival and spread in the hospital environments. On the other hand, similar resistance profiles were observed among MDR isolates of different PFGE types (Table 3). One possible factor responsible for the emergence of similar resistance patterns among these isolates could be due to selective pressure from extensive use of antimicrobial agents.³

Analysis of integrase genes showed that approximately 54% of our isolates (45/83) carried class 1 integrase genes, whereas only two isolates contained a class 2 integrase gene and none of the isolates carried a class 3 integrase gene (Table 3). These results were comparable to those from a previous study in the UK.²² Both of our class 2 integrase gene-carrying isolates (Aci19 and Aci20) contained the same class 2 gene cassette arrays and were recovered from the patients around the same period (Table 3, PFGE types 19 and 6b). This led to the speculation that the two genetically distinct strains might have acquired class 2 integrons through horizontal gene transfer. Additional analysis showed that 44 of 46 isolates containing integrase genes (∼96%) exhibited resistance to ≥2 antimicrobial agents tested (Table 1, antibiogram types 6–21). This finding is consistent with the previous studies in that the majority of strains carrying integrons exhibited multiple drug resistance.^14,22,32 Nevertheless, 18 MDR isolates (15 belonging to predominant PFGE type 36) did not carry any of the three classes of integrase genes (Table 3).

In this study, 9 out of 13 MDR isolates belonging to the predominant PFGE type 7 carried the 2.4-kb class 1 gene cassette array consisting of aacA4-catB8-aadA1 (Table 3). The association of this 2.4-kb class 1 gene cassette array with multiresistant outbreak clones has previously been reported in the United Kingdom,⁴⁶ Taiwan,^7,25 and China.^50,52 Moreover, unlike the results obtained by Xu et al.,⁵² we observed a correlation between strains carrying this gene cassette array and resistance to AMK (Table 3). The 2.5- and 3-kb class 1 gene cassette arrays revealed that the two arrays, consisting of aacC1-orfX-orfX′-aadA1a and aacC1-orfX-orfX-orfX′-aadA1a, respectively, were identical to those found in outbreak clones from the United Kingdom.⁴⁶ The 2.5-kb array has also been found in A. baumannii strains from many countries in Asia^18,25 and Europe,^15,24,30,53 as well as an MDR clinical isolate from a horse.¹ Aside from this study, the 1.2-kb class 1 gene cassette array containing dfrA1-orfC was also found in A. baumannii strain AYE, in multidrug-resistant Salmonella enterica strains isolated from seafood^20,21 and in clinical strains of Vibrio cholerae from India.⁴⁰ Identification of gene cassettes on A. baumannii integrons revealed not only antimicrobial resistance genes but also unknown open reading frames (orfs) like orfX, orfX′, and orfC.^{17,24,25,38,46} Since most antimicrobial resistance genes found on integrons usually confer resistance to certain aminoglycosides and β-lactam agents, it was speculated that proteins encoded by these orfs may make some contribution to the epidemicity of strains carrying integrons.⁴⁶ Although in our study, strains harboring the unknown orfs located on the 1.2-, 2.5-, or 3-kb class 1 gene cassette arrays appeared to be genetically unrelated (Table 3), the epidemic potential of these isolates should not be overlooked.

The presence of the 1.5-kb class 1 gene cassette array containing bla_IMP-14-aac(6′) has so far been described only in clinical isolates of Pseudomonas aeruginosa (GenBank accession no. AY553332) and A. baumannii (GenBank accession no. GQ302618) from Thailand. Our bla_IMP-14-positive isolate, Aci16, was susceptible to IPM, and this was not surprising because Acinetobacter isolates carrying MBL genes that appeared to be susceptible to impenem were previously reported.⁴⁹ Further investigation to detect expression of MBLs in this strain is in progress. Other than the production of class B family of MBLs, carbapenem resistance in A. baumannii can also be due to the production of the OXA-type class D family of serine β-lactamases.^5,49 We did not detect integrons carrying carbapenemase resistance genes in any of our IPM-resistant isolates. Thus, in our case, it is very likely that the carbapenemase resistance genes may not be associated with integrons.

In conclusion, results from this study indicated the dissemination of the three MDR clones in Ramathibodi hospital. This necessitates the need for further investigations to identify a reservoir for the organism, as well as individual risk factors and modes of transmission to help in the implementation of effective infection control measures. In addition, all the identified gene cassette arrays located on the class 1 and class 2 integrons were identical to those previously identified from Europe and Asia with the exception of the 1.5-kb integron carrying the variant IMP-14 that seemed to be restricted to Thai isolates of Pseudomonas aeruginosa and A. baumannii.³⁷

Footnotes

Acknowledgments

This work was supported by a target grant from Mahidol University, Thailand, to the Center of Emerging Bacterial Infections. We thank Dr. James Dubbs for editing the article.

Disclosure Statement

No competing financial interests exist.

References

Abbott

, O'Mahony

, Leonard

, Quinn

P.J.

, van der Reijden

, Dijkshoorn

et al. 2005. Characterization of a 2.6 kbp variable region within a class 1 integron found in an Acinetobacter baumannii strain isolated from a horse. J. Antimicrob. Chemother, 55:367–370.

Aswapokee

, Charoensook

, Sangsiriwut

1998. Antimicrobial resistance pattern of Acinetobacter spp. J. Infect. Dis. Antimicrob. Agents, 15:43–48.

Ayuthya

S.K.

, Matangkasombut

O.P.

, Sirinavin

, Malathum

, Sathapatayavongs

2003. Utilization of restricted antibiotics in a university hospital in Thailand. Southeast Asian J. Trop. Med. Public Health, 34:179–186.

Bartual

S.G.

, Seifert

, Hippler

, Luzon

M.A.

, Wisplinghoff

, Rodriguez-Valera

2005. Development of a multilocus sequence typing scheme for characterization of clinical isolates of Acinetobacter baumannii. J. Clin. Microbiol., 43:4382–4390.

Brown

, Amyes

2006. OXA (beta)-lactamases in Acinetobacter: the story so far. J. Antimicrob. Chemother., 57:1–3.

Chaiwarith

, Mahatthanaphak

, Boonchoo

, Supparatpinyo

, Sirisanthana

2005. Pandrug-resistant Acinetobacter baumannii at Maharaj Nakorn Chiangmai Hospital. J. Infect. Dis. Antimicrob. Agents, 22:1–8.

Chen

Y.S.

, Lin

H.H.

, Wu

C.H.

, Hsiao

Y.S.

, Hsu

N.S.

, Chen

Y.L.

2009. Colonization of a medical center in Southern Taiwan by epidemic strains of carbapenem- and multidrug-resistant Acinetobacter baumannii and the genetic organization of their integrons. Jpn. J. Infect. Dis., 62:155–157.

Clinical Laboratory Standards Institute. 2006. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard. 7th CLSI document M7-A7. Clinical Laboratory Standards Institute: Wayne, PA.

Diancourt

, Passet

, Nemec

, Dijkshoorn

, Brisse

The population structure of Acinetobacter baumannii: expanding multiresistant clones from an ancestral susceptible genetic pool. PLoS One, 5:e10034.

10.

Dijkshoorn

, Aucken

, Gerner-Smidt

, Janssen

, Kaufmann

M.E.

, Garaizar

et al. 1996. Comparison of outbreak and nonoutbreak Acinetobacter baumannii strains by genotypic and phenotypic methods. J. Clin. Microbiol, 34:1519–1525.

11.

Dillon

, Thomas

, Mohmand

, Zelynski

, Iredell

2005. Multiplex PCR for screening of integrons in bacterial lysates. J. Microbiol. Methods, 62:221–232.

12.

Fournier

P.E.

, Richet

2006. The epidemiology and control of Acinetobacter baumannii in health care facilities. Clin. Infect. Dis., 42:692–699.

13.

Gallego

, Towner

K.J.

2001. Carriage of class 1 integrons and antibiotic resistance in clinical isolates of Acinetobacter baumannii from northern Spain. J. Med. Microbiol., 50:71–77.

14.

Gaur

, Prakash

, Anupurba

, Mohapatra

T.M.

2007. Possible role of integrase gene polymerase chain reaction as an epidemiological marker: study of multidrug-resistant Acinetobacter baumannii isolated from nosocomial infections. Inter. J. Antimicrob. Agents, 29:446–450.

15.

Gombac

, Riccio

M.L.

, Rossolini

G.M.

, Lagatolla

, Tonin

, Monti-Bragadin

et al. 2002. Molecular characterization of integrons in epidemiologically unrelated clinical isolates of Acinetobacter baumannii from Italian hospitals reveals a limited diversity of gene cassette arrays. Antimicrob. Agents Chemother, 46:3665–3668.

16.

Gonzalez

, Sossa

, Bello

, Dominguez

, Mella

, Zemelman

1998. Presence of integrons in isolates of different biotypes of Acinetobacter baumannii from Chilean hospitals. FEMS Microbiol. Lett., 161:125–128.

17.

, Tong

, Zhao

, Liu

, Ning

, Pan

et al. 2007. Prevalence and characterization of class I integrons among Pseudomonas aeruginosa and Acinetobacter baumannii isolates from patients in Nanjing, China. J. Clin. Microbiol, 45:241–243.

18.

Hwa

W.E.

, Subramaniam

, Navaratnam

, Sekaran

S.D.

2009. Detection and characterization of class 1 integrons among carbapenem-resistant isolates of Acinetobacter spp. in Malaysia. J. Microbiol. Immunol. Infect., 42:54–62.

19.

Jamulitrat

, Thongpiyapoom

, Suwalak

2007. An outbreak of imipenem-resistant Acinetobacter baumannii at Songklanagarind Hospital: the risk factors and patient prognosis. J. Med. Assoc. Thai., 90:2181–2191.

20.

Khan

A.A.

, Cheng

C.M.

, Van

K.T.

, West

C.S.

, Nawaz

M.S.

, Khan

S.A.

2006. Characterization of class 1 integron resistance gene cassettes in Salmonella enterica serovars Oslo and Bareily from imported seafood. J. Antimicrob. Chemother., 58:1308–1310.

21.

Khan

A.A.

, Ponce

, Nawaz

M.S.

, Cheng

C.M.

, Khan

J.A.

, West

C.S.

2009. Identification and characterization of Class 1 integron resistance gene cassettes among Salmonella strains isolated from imported seafood. Appl. Environ. Microbiol., 75:1192–1196.

22.

Koeleman

J.G.

, Stoof

, van der Bijl

M.W.

, Vandenbroucke-Grauls

C.M.

, Savelkoul

P.H.

2001. Identification of epidemic strains of Acinetobacter baumannii by integrase gene PCR. J. Clin. Microbiol., 39:8–13.

23.

Koomanachai

, Tiengrim

, Kiratisin

, Thamlikitkul

2007. Efficacy and safety of colistin (colistimethate sodium) for therapy of infections caused by multidrug-resistant Pseudomonas aeruginosa and Acinetobacter baumannii in Siriraj Hospital, Bangkok, Thailand. Int. J. Infect. Dis., 11:402–406.

24.

Kraniotaki

, Manganelli

, Platsouka

, Grossato

, Paniara

, Palu

2006. Molecular investigation of an outbreak of multidrug-resistant Acinetobacter baumannii, with characterisation of class 1 integrons. Inter. J. Antimicrob. Agents, 28:193–199.

25.

Lee

Y.-T.

, Huang

L.-Y.

, Chen

T.-L.

, Siu

L.-K.

, Fung

C.-P.

, Cho

W.-L.

et al. 2009. Gene cassette arrays, antibiotic susceptibilities, and clinical characteristics of Acinetobacter baumannii bacteremic strains harboring class 1 integrons. J. Microbiol. Immunol. Infect, 42:210–219.

26.

Levesque

, Piche

, Larose

, Roy

P.H.

1995. PCR mapping of integrons reveals several novel combinations of resistance genes. Antimicrob. Agents Chemother., 39:185–191.

27.

Mazel

2006. Integrons: agents of bacterial evolution. Nat. Rev. Microbiol., 4:608–620.

28.

Mazel

, Dychinco

, Webb

V.A.

, Davies

2000. Antibiotic resistance in the ECOR collection: integrons and identification of a novel aad gene. Antimicrob. Agents Chemother., 44:1568–1574.

29.

Murray

P.R.

2005. Manual of Clinical Microbiology, 8th. American Society for Microbiology: Washington, DC.

30.

Nemec

, Dolzani

, Brisse

, van den Broek

, Dijkshoorn

2004. Diversity of aminoglycoside-resistance genes and their association with class 1 integrons among strains of pan-European Acinetobacter baumannii clones. J. Med. Microbiol., 53:1233–1240.

31.

Niumsup

P.R.

, Boonkerd

, Tansawai

, Tiloklurs

2009. Carbapenem-resistant Acinetobacter baumannii producing OXA-23 in Thailand. Jpn. J. Infect. Dis., 62:152–154.

32.

J.Y.

, Kim

K.S.

, Jeong

Y.W.

, Cho

J.W.

, Park

J. C.

, Lee

J.C.

2002. Epidemiological typing and prevalence of integrons in multiresistant Acinetobacter strains. APMIS., 110:247–252.

33.

Pachon-Ibanez

M.E.

, Jimenez-Mejias

M.E.

, Pichardo

, Llanos

A.C.

, Pachon

2004. Activity of tigecycline (GAR-936) against Acinetobacter baumannii strains, including those resistant to imipenem. Antimicrob. Agents Chemother., 48:4479–4481.

34.

Peleg

A.Y.

, Seifert

, Paterson

D.L.

2008. Acinetobacter baumannii: emergence of a successful pathogen. Clin. Microbiol. Rev., 21:538–582.

35.

Perez

, Hujer

A.M.

, Hujer

K.M.

, Decker

B.K.

, Rather

P.N.

, Bonomo

R.A.

2007. Global challenge of multidrug-resistant Acinetobacter baumannii. Antimicrob. Agents Chemother., 51:3471–3484.

36.

Rungruanghiranya

, Somboonwit

, Kanchanapoom

2005. Acinetobacter infection in the intensive care unit. J. Infect. Dis. Antimicrob. Agents, 22:77–92.

37.

Samuelsen

, Toleman

M.A.

, Sundsfjord

, Rydberg

, Leegaard

T.M.

, Walder

et al. 2010. Molecular epidemiology of metallo-beta-lactamase-producing Pseudomonas aeruginosa isolates from Norway and Sweden shows import of international clones and local clonal expansion. Antimicrob. Agents Chemother, 54:346–352.

38.

Segal

, Thomas

, Gay Elisha

2003. Characterization of class 1 integron resistance gene cassettes and the identification of a novel IS-like element in Acinetobacter baumannii. Plasmid, 49:169–178.

39.

Seifert

, Stefanik

, Wisplinghoff

2006. Comparative in vitro activities of tigecycline and 11 other antimicrobial agents against 215 epidemiologically defined multidrug-resistant Acinetobacter baumannii isolates. J. Antimicrob. Chemother., 58:1099–1100.

40.

Shi

, Fujihara

, Sato

, Ito

, Garg

, Chakrabarty

et al. 2006. Distribution and characterization of integrons in various serogroups of Vibrio cholerae strains isolated from diarrhoeal patients between 1992 and 2000 in Kolkata, India. J. Med. Microbiol, 55:575–583.

41.

Stokes

H.W.

, Hall

R.M.

1989. A novel family of potentially mobile DNA elements encoding site-specific gene-integration functions: integrons. Mol. Microbiol., 3:1669–1683.

42.

Tenover

F.C.

, Arbeit

R.D.

, Goering

R.V.

, Mickelsen

P.A.

, Murray

B.E.

, Persing

D.H.

et al. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol, 33:2233–2239.

43.

Thongpiyapoom

, Narong

M.N.

, Suwalak

, Jamulitrat

, Intaraksa

, Boonrat

et al. 2004. Device-associated infections and patterns of antimicrobial resistance in a medical-surgical intensive care unit in a university hospital in Thailand. J. Med. Assoc. Thai, 87:819–824.

44.

Tiengrim

, Tribuddharat

, Thamlikitkul

2006. In vitro activity of tigecycline against clinical isolates of multidrug-resistant Acinetobacter baumannii in Siriraj hospital, Thailand. J. Med. Assoc. Thai., 89:S102–S105.

45.

Tribuddharat

, Tiensasiton

, Techachaiwiwat

, Rugdeekha

, Dhiraputtra

, Thamlikitkul

2003. In vitro activity of polymyxin E against multi-drug resistant Pseudomonas aeruginosa and Acinetobacter baumannii. J. Infect. Dis. Antimicrob. Agents, 20:135–137.

46.

Turton

J.F.

, Kaufmann

M.E.

, Glover

, Coelho

J. M.

, Warner

, Pike

et al. 2005. Detection and typing of integrons in epidemic strains of Acinetobacter baumannii found in the United Kingdom. J. Clin. Microbiol, 43:3074–3082.

47.

Turton

J.F.

, Woodford

, Glover

, Yarde

, Kaufmann

M.E.

, Pitt

T.L.

2006. Identification of Acinetobacter baumannii by detection of the blaOXA-51-like carbapenemase gene intrinsic to this species. J. Clin. Microbiol., 44:2974–2976.

48.

van den Broek

P.J.

, van der Reijden

T.J.

, van Strijen

, Helmig-Schurter

A.V.

, Bernards

A.T.

, Dijkshoorn

2009. Endemic and epidemic Acinetobacter species in a university hospital: an 8-year survey. J. Clin. Microbiol., 47:3593–3599.

49.

Walsh

T.R.

, Toleman

M.A.

, Poirel

, Nordmann

2005. Metallo-beta-lactamases: the quiet before the storm? Clin. Microbiol. Rev., 18:306–325.

50.

Wang

, Guo

, Sun

, Wang

, Yang

, Chen

et al. 2007. Molecular epidemiology of clinical isolates of carbapenem-resistant Acinetobacter spp. from Chinese hospitals. Antimicrob. Agents Chemother, 51:4022–4028.

51.

White

P.A.

, McIver

C.J.

2001. Integrons and gene cassettes in the Enterobacteriaceae. Antimicrob. Agents Chemother., 45:2658–2561.

52.

, Kong

, Cheng

, Yan

, Du

, Gai

et al. 2008. Integron gene cassettes in Acinetobacter spp. strains from South China. Int. J. Antimicrob. Agents, 32:441–445.

53.

Zarrilli

, Crispino

, Bagattini

, Barretta

, Di Popolo

, Triassi

et al. 2004. Molecular epidemiology of sequential outbreaks of Acinetobacter baumannii in an intensive care unit shows the emergence of carbapenem resistance. J. Clin. Microbiol, 42:946–953.