The increasing significance of outbreaks of Acinetobacter spp.: the need for control and new agents

journal

of Hospital

OUTBREAKS

Infection

(1995)

30 (Supplement),

441-452

CAUSED BY BACTERIA WITH NOVEL TOWARDS ZERO THERAPEUTIC

The increasing significance Acinetobacter spp.: the need agents

MULTIPLE OPTIONS?

RESISTANCE:

of outbreaks of for control and new

E. Bergogne-Bc!x+zin Department of Microbiology, Bichat-Claude Bernard University-Hospital, Rue Henri-Huchard, 75877 Paris Cedex 18, France

46

Summary:

Acinetobacter spp. are Gram-negative non-fermentative bacteria which may be isolated as commensals from human skin, throat and intestine but are also increasingly responsible for hospital infections. Owing to frequent changes in their taxonomy, their pathogenic role in humans has not been clear but today acinetobacter is considered to be a significant nosocomial pathogen in outbreaks of hospital infections predominantly in intensive care units. Nosocomial infections due to acinetobacter include urinary tract infections, bacteraemia, wound and burn infections, but also they are frequently isolated from ventilator-associated nosocomial pneumonia. The frequency of hospital outbreaks of acinetobacter infections has required the development of reliable typing methods. As well as conventional ‘phenotypic’ methods (serology, biotyping, phage typing), ‘genotypic’ systems (ribotyping, plasmid profiles, pulsed-field gel electrophoresis) have been utilized for strain identification. These typing systems should allow a better understanding of the epidemiology of acinetobacter in the hospital environment, e.g. sources, modes of transmission, and result in more efficient preventive measures. Acinetobacter infections are difficult to treat owing to their frequent multiple resistance to the antibiotics currently available for the treatment of nosocomial infections; various mechanisms of resistance to p-lactams and aminoglycosides have been identified in the genus. Combination therapy is usually recommended for treatment of acinetobacter nosocomial infections and active antibacterials include imipenem, ceftazidime, amikacin and the newer fluoroquinolones. Careful in vitro testing of the activity of combinations of these drugs is recommended prior to their use. Keywords: demiological

Acinetobacter markers.

spp.;

nosocomial

infections;

outbreaks;

epi-

Introduction Acinetobacter strains are non-fermentative, non-fastidious, aerobic Gramnegative coccobacillary organisms, usually demonstrating diploid formation, or in chains of variable length. They are commonly present in soil and water as free-living saprophytes and are also isolated as commensals from skin, throat and various sites in healthy people. There have been frequent 0195-6701/95/060441+

62 1995 The Hospital

12 $08.00/O

441

Infection

Soaety

442

E. Bergogne-BBrBzin

changes in their taxonomy so that their pathogenic role in humans has only recently been understood. Recent genetic studies have clarified their taxonomy’ and many epidemiological investigations have been carried out.‘-’ As a nosocomial pathogen, acinetobacter is an important cause of morbidity and mortality particularly in debilitated patients in intensive care unit (ICU) settings. Many outbreaks of infection due to A. baumannii have been described and these have usually been associated with colonization of equipment used in respiratory therapy. Hand carriage of the organism by hospital personnel has also been frequently implicated in epidemic transmission.’ This review focuses on the epidemiological features of acinetobacter nosocomial infections and examines the available epidemiological markers for delineation of outbreaks as well as preventive and therapeutic measures for control of outbreaks.

Epidemiology

Environmental sources Acinetobacter strains are ubiquitous organisms which are widely distributed in nature and may be found in fresh water samples when appropriate culture techniques are used.“8 Acinetobacter spp. can also be isolated from food and animals and forms part of the normal flora of fresh meats. In hospitals, the organism is frequently present in the inanimate environment in moist situations such as room humidifiers, tap water, sinks, and in all types of ventilatory equipment2 which form aerosols. A seasonal incidence of acinetobacter infections, with an increase in the rate of infection in late summer and early winter, may be related to temperature and humidity.’ Human carriage Acinetobacters have been found in the bacterial flora of the skin, in the axillae, groin and toe webs of normal individuals.7’10,” Less frequently, the organism can colonize the normal intestine but this does not seem to constitute a negligible reservoir.r2 The skin of patients and staff (hand carriage) have been implicated in the spread of the organism in outbreaks of acinetobacter infections.7”3 Pathogenic role of Acinetobacter spp. Risk factors predisposing to severe infections include severe underlying disease, malignancy, burns, immunosuppression, major surgery and age particularly in the elderly and neonates.14 Other risk factors which have been identified in outbreaks of acinetobacter infection are prolonged stay in ICU, prolonged ventilatory support and other mechanical devices, and wide use of broad-spectrum antibiotics.“’ In French hospitals, the isolation rate of acinetobacter from nosocomial infections has recently been reported

Outbreaks Table

I.

Distribution us

of Acinetobacter

(21)

origin

Others Catheters (Central nervous system, intraabdominal, burns, cardiovascular system, etc.) * Paediatric [ 1, Number

University-Hospital Bichat-Claude

443

by sites of infection

(89)

(28)”

Belgium

OW

1981

1991

1991

ii.9

32 17.5

21 27

30.5 25.6

21.5 9.3

34 6

27.5 7.5

19.1 (f16) 7.5 (rfr5)

22.3 7.6

13.3 -

37

15.5 2

12(f7) 5 (5)

18.1 -

intensive care unit (ICU): of strains; NS, not stated;

15%; b urn ( ), number

units: 8%; of centres.

medical,

surgical

(55)

% [237]

% (&SD)

Bernard

1974-1977

Urine Tracheobronchial specimens Pus wounds Blood cultures

spp.

strains France

NNIS (%) [1372] Clinical

of Acinetobacter

1990-1991

(-&15) (f13)

ICUs:

27 24.8

77%.

as 9.7% and acinetobacter isolates represent 15.6% of the total Gramnegative bacillary isolates recovered from nosocomial pneumonia in the country.4 The distribution by body site of acinetobacter infections does not appear to differ from that of other nosocomial Gram-negative bacteria. Table I shows that the main sites of infection in several surveys5,‘4p15were the lower respiratory tract and the urinary tract, with rates of infection ranging from 1530% of total acinetobacter infections. Large outbreaks of pneumonia have been described in patients with severe underlying disease requiring assisted ventilation; most were intubated. Outbreaks

of hospital

infections

due

to acinetobacter

A number of outbreaks in hospitals have been described over the last 20 years as microbiologists and physicians have become aware of the increasing importance of Acin&tobacter spp. as nosocomial pathogens.3’13-‘6 Table II summarizes many of the outbreaks described in the medical literature from 1977-1994. Various typing methods have been devised for epidemiological investigation of the genus and have been applied to many hospital outbreaks. They can conveniently be separated into phenotypic and genotypic systems. Each have their advantages in terms of sensitivity and reproducibility and their use for the identification of strains is provided in Table III. Phenotypic typing systems Traditional systems were based mainly on phenotypic characters of the bacteria, such as identification and biotyping by means of assimilation tests,

E. Bergogne-BBrC?zin

444 Table

II.

Summary

of several

outbreaks

Acinetobacter species (No. of isolates)*

Infection/ colonization

A. calcoaceticus (1372)

Endemic cases 29% respiratory tract 27% urine 21.5% wound 15 episodes of septicaemia

A. anitratus

(15)

A. calcoaceticus

Acinetobacter (15)

A. anitratus (epidemic

(62)

spp.

(1) strain)

A. calcoaceticus subsp. anitratus Acinetobacter

(26)

spp.

(16) 13 A. baumannii 2 A. haemolyticus 1 A. 1wofJii Acinetobacter spp. (2685)

Acinetobacter calcoaceticus var. anitratus Acinetobacter

(237)

(40) spp.

site

Endemic cases 6 pus 35 urine 4 faeces 11 respiratory tract other: environment Hospital outbreaks 7 respiratory tract 2 blood 3 miscellaneous 2 environment Hospital outbreaks (40 patients) urine skin environment Endemic cases 25 patients pneumonia Hospital outbreaks 68% throat 25% rectum 31.5% both (newborns) Endemic cases 27% pneumonia 21% urine 7% septicaemia Hospital outbreaks 90% pneumonia 10% bacteraemia Outbreaks (285 episodes) 18 blood 59 respiratory tract 53 skin/wound 64 urine 43 other

OY endemic Typing systems

cases of Acinetobacter

infections

References used

Antibiotyping

Glew

et al.,

1977

Antibiotyping Serotyping (28 serotypes; 50% untypable strains)+ Phage typing (79% typable, 18 &age types)

Ramphal

Vieu

et al.,

1979

Phage typing (phage type predominant)

Vieu

et al.,

1980

et al.,

1979

79

Antibiotyping Biotyping

French

Antibiotyping Biotyping Plasmid analysis Antibiotyping Biotyping Phage typing (75% phage type

Vila

et al.,

et al.,

1980

1989

Personal data, unpublished, 1987 97)

Antibiotyping Phage typing: major types: 17 (31%), 124 (38%) Antibiotyping

Joly-Guillou 1992

Biotyping Antibiotyping Macrorestriction analysis PCR fingerprint

Struelens 1993

Urban

et al.,

et al..

1993

et al.,

(continued)

Outbreaks Table Acinetobacter species (No. of isolates)*

Infection/ colonization

A. baumannii

(28)

A. baumannii (out of 400)

(103)

Endemic cases septicaemia pneumonia peritonitis (APD) 9 outbreaks

* Taxonomy

of isolates

0, Number trophoresis.

of types

phage typing, analysis.

is cited

II.

site

serotyping,

APD,

ambulatory

445

(continued)

in the table as presented

identified;

spp.

of Acinetobacter

Typing systems

References used

Biotyping (25% me 9) Ribotyping (N 78%: 4 ribotypes) 4 typing systems Biotyping (4) Plasmid analysis (6) Antibiotyping (5) PFGE (8) in corresponding peritoneal dialysis;

and antibiogram

Gerner-Smidt,

Seifert

et al.,

articles. PFGE, pulsed-field

1992

1994

gel

elec-

typing by resistance phenotype

Biotyping. The genus Acinetobacter comprises 12 DNA groups defined by DNA-DNA hybridization.* Phenotypic tests can be used to speciate isolates and 17 phenotype characters have been proposed, based on carbon source utilization tests, gelatin hydrolysis and a series of other substrate utilization tests.i7 Table IV lists the species designation, DNA group and epidemiological features of 12 Acinetobacter spp. Biotyping has been used as an epidemiological marker in many recent investigations5,‘8”9 to type nosocomial strains of A. baumannii, but needs to be used in conjunction with other typing systems owing to its poor discrimination. Antibiogram typing. The antibiotic resistance pattern has been widely used as a marker5,16’20to differentiate outbreak strains from sporadic strains and the p-lactam resistance profile is a useful indicator of acinetobacter cross-infection.21 The main limitation of this method as an epidemiological marker is its lack of specificity, as several strains with the same resistance pattern may belong to different strain types owing to acquisition of the same transposable elements or plasmids.4’22 Many strains possess a capsule which has been used for typing by an immunofluorescent antibody technique: 28 serotypes were recognizable with this method, but there is little experience of its application in a clinical setting.’ Another method, with different rabbit antisera, has been used which identified 20 serovars23 by checkerboard tube agglutination. Several outbreaks of nosocomial cross-infection caused by serovars 4 and 10 were delineated. Serotyping seems to be a satisfactory marker but as it is based on surface structures of acinetobacter it may not be a stable characterz4 and thus this system may fail in the typing of strains. Serotyping.

E. Bergogne-BCrCzin

446 Table

Epidemiological

III.

Typing

systems

Phenotypic Biotyping

Serotyping

Bacteriocin

typing

typing

system

Newer epidemiologic Cell envelope protein patterns

Molecular Ribotyping

biology

Plasmid

typing

Plasmid fingerprinting PFGET

PCR$

of

acinetobacter

Comments

Reference

17 biotypes: identification method poorly discriminative Definition of resistance phenotypes May identify cross-infection Lack of specificity Acquisition or loss of plasmids/ transposons Immunofluorescent antibody technique: 28 serotypes Polyclonal immune rabbit sera: 20 serovars Based on surface structures (capsule): unstable Gillies and Govan procedure (1966) Two commonest types a, c 75% of A. anitratus were bacteriocin producers 65% were typable Poorly discriminative 21 phage types since 1976 + a second set of 14 phages Definition of 125 types and 25 subtypes Predominant phage types 17 and 34 in outbreaks Moderately discriminative

Bouvet

outbreaks

systems

Antibiogram

Phage

markers available for investigation

fingerprints

SDS-PAGE, electrophoresis;

sodium PCR,

markers SDS-PAGE technique 30-50 protein bands Four common patterns in outbreak strains Discriminative but intrastrain variation of SDS-PAGE patterns techniques Restriction enzymes EcoRI and Hind111 or C&I, Sol1 High discriminative power Correlate with SDS-PAGE Plasmid patterns Common to strains with similar resistance pattern 2-6 different plasmid bands Poor resolution Plasmids unstable 27 different plasmid profiles in 86.5% of 37 outbreak strains Highly discriminative Chromosomal DNA Macrorestriction patterns by using endonucleases NheI and SmaI Good discrimination of epidemic related strains dodecyl sulphate-polyacrylamide polymerase chain reaction.

et al., 1987 et al.,

Bergogne-BCrtzin

1987 Joly-Guillou Traub et

et al., 1988 aZ., 1989

Ramphal Traub,

et al., 1979 1989

Andrews,

1986

Vieu et al., Bergogne-Btrezin 1987 Lambert et

1980

Dijkshoorn

et al., 1993

Dijkshoorn Gerner-Smidt,

et al., 1993

Vila et al., Joly-Guillou (unpublished)

al., 1990

1992

1989

et al., 1994

Seifert

Struelens,

gel electrophoresis;

et al.,

PFGE,

1993

pulsed-field

gel

Outbreaks Table

IV.

Microbiological

Genomic Species

A.

designation

calcoaceticus

and

species

1

Unnamed

3 4

A. junii Unnamed A. johnsonii

A.

8

lwoffii

of Acinetobacter

* Limited nutritional t Glucose

taxonomic abilities positive:

species

Epidemiological features

Glucose growth Glucose growth

positivet; at 41°C positive; at 44°C

no

Glucose positive; growth at 41°C Haemolytic strains and 96% gelatin positive; glucose positive (52%) Glucose negative Glucose hydrolysis Glucose

+ (50%);

Glucose

negative

Glucose Glucose Glucose

positive negative negative

gelatin

negative

according

and

to

clinical

10 11 12

significance of glucose according to reference 1). acidified.

oxidation

Presence

in soil

Hospital environment; often isolated from humans; associated with nosocomial infections Soil; clinical samples Hospital environment; occasionally in patients Clinical samples; environment Environment; animals; skin of staff in hospitals; clinical specimens Animals (chicken); clinic:1 samples; hands of hospital staff Associated with cotton; clinical specimens

9 Unnamed Unnamed A. radioresistens

447

DNA group

2

ticus

characteristics new taxonomy’

spp.

Main physiological features*

A. baumannii

A. haemoly

clinical

of Acinetobacter

(better

discrimination

of

species

by

using

Other phenotypic methods. A bacteriocin typing method has also been described and has been shown to be useful in epidemiological investigations.25 A phage typing system, developed at the Phage Typing Centre of the Pasteur Institute,24 was based on the use of two complementary series of bacteriophages isolated from sewage, and which are highly specific for A. calcoaceticus. A first set of 21 phages used since 1976 for routine investigation of outbreaks of acinetobacter infections, complemented by a second set of 14 phages, has been used to analyse about 2500 clinical isolates from France and other European countries. The predominant phage types recognized in outbreaks were 17 and 24.22,26 Modern approaches to epidemiological studies with acinetobacter The development of molecular biology techniques and their application has provided new epidemiological markers for surveys of acinetobacter outbreaks.

448

E. Bergogne-BCrCzin

Cell envelope protein patterns. Typing of A. baumannii has been proposed on the basis of the electrophoretic patterns of cell-envelope proteins.18’27 The technique of sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) of cell-envelope proteins has been useful in distinguishing between bacterial strains and in studying the dissemination of acinetobacter strains within the hospital. Electrophoretic patterns of cell-envelope proteins are characterized by 30-50 protein bands of different staining intensities. Most isolates within an outbreak are usually very similar in cell-envelope protein and this reduces the discriminating potential of the method. Cellenvelope profiles are, however, stable and reproducible for acinetobacter, and variation of patterns within patients may be connected with the concurrent presence of different strains.‘* Ribotyping. A new molecular typing method has been applied to epidemiological studies of acinetobacter. This method is based on DNA extraction and digestion with restriction enzymes EcoRI and HindIII” or CZaI and SaZI,4 separation of restriction fragments in agarose gels by electrophoresis, and Southern transfer and hybridization with a biotinylated cDNA probe prepared from Escherichia coli rRNA. Within each outbreak analysed in the study, the isolates were uniform in their banding pattern and a remarkable degree of correlation was found with SDS-PAGE. The efficacy of ribotyping for identification of outbreaks of acinetobacter infections has been demonstrated in these studies and these systems might be used in daily clinical microbiological practice to provide a reliable typing system with a high discriminatory power. The overall agreement between biotyping, cell-envelope electrophoresis, ribotyping and comparison of antibiograms has been experienced in recent studies.5,18 Biotyping and antibiogram typing were the least discriminatory. Strain differentiation was considered acceptable with electrophoresis or ribotyping, although it was concluded that verification by another method is needed before deciding that apparently indistinguishable isolates belong to the same strain.” Analysis of chromosomal DNA by pulsed-$eld gel electrophoresis (PFGE) and chromosomal DNA macro-restriction patterns. These have been assayed recently, using the endonucleases NheI and SmaI:’ these methods were used as markers for strain identity, based on genetic similarities between strains. Genotyping by the polymeruse chain reaction (PCR) . PCR typing was also performed in the above study’ and resulted in satisfactory discrimination between epidemic strains and control strains isolated from patients with sporadic infection in the same hospital. The authors concluded that DNA polymorphism within the species A. baumannii can be detected by macrorestriction analysis by PFGE. PCR typing assays had lower discrimination between acinetobacter strains of the same biotype but there was general agreement overall between the PCR fingerprinting method and DNA

Outbreaks

of Acinetobacter

spp.

449

macrorestriction analysis as epidemiological markers. Both of these techniques have been applied in two outbreaks of acinetobacter hospital infections’ and seem to be promising tools for outbreak delineation and detection of micro-epidemics. Plasmid analysis. Vila et a1.16determined the plasmid profiles of 25 acinetobacter isolates in a hospital outbreak and showed two patterns characterized by one and two plasmids, respectively. The same plasmid pattern was common to all strains with similar antibiograms. A study in progress has shown, in 15 acinetobacter isolates, the presence of six different plasmid profiles (M. L. Joly-Guillou, personal communication): strains showed one to six different bands of molecular weights between 80 and 15 MDa. In another recent study2’ plasmid analysis revealed six different and two related patterns, but one outbreak strain lacked plasmids. Therefore, although plasmid profile analysis provides a cost-effective first step in epidemiological typing of A. baumannii, plasmids are unstable genetic structures and are not present in all strains. Thus plasmid profiling is not the most reliable of markers. Acinetobacter

and

antibiotics

Antibiotic susceptibility/resistance of Acinetobacter spp. In early in vitro studies, the majority of acinetobacter clinical isolates were susceptible to ampicillin, cephalosporins, minocycline;‘3S15,29 in 1969, only 3.4% were resistant to carbenicillin and most acinetobacter isolates were susceptible to gentamicin. High proportions of strains have become resistant to antibiotics, and at the present time some acinetobacter strains are resistant to most commonly used antibacterial drugs, including aminopenicillins, ureidopenicillins, first and second generation cephalosporins, cephamycins and tet(cefoxitin),21 aminoglycosides-aminocychtols,22~29 chloramphenicol racyclines. Differences in susceptibilities are noted between countries due to environmental factors, and different patterns of antibiotic usage that exist in Japan, in Germany,30 Spain16 or France.21Z29Indeed, aminoglycosides now seem to be inactive against A. baumannii in Germany.30 With some such as third generation cephalosporins (cefotaxime, major antibiotics, ceftazidime), imipenem, tobramycin and amikacin, the minimal inhibitory concentrations (MICs) have increased between 1980 and 1992. Acinetobacter strains, initially susceptible to fluoroquinolones, have become resistant to pefloxacin within five years with the proportions of resistant isolates rising from approximately 20-80%. Imipenem remains the most active drug but imipenem resistance, although still limited (O-5-5%), is a threat for the near future. A. Zwoffii strains are more susceptible to plactams than A. baumannii, and A. haemolyticus is highly resistant to aminoglycosides. Rifampicin has been tested29 and the mean MICs for A. baumannii are currently 2-4 mg L-‘. Even though this drug is not normally

450

E. Bergogne-BCrCzin

considered to be active against the genus, its empirical use in combinations with imipenem often provides synergistic effects useful for the treatment of patients in intensive care units in France (unpublished data); A. haemolyticus is highly resistant to rifampicin. Other species, A. Zwojjii, A. johnsonii and A. junii, isolated from the hospital environment, are less frequently involved in nosocomial infections and they are generally less resistant to antibiotics. Therapeutic problems of nosocomial infections due to acinetobacter Multiply-resistant A. baumannii outbreaks constitute one of the major problems in ICUs. Only a limited number of major antibiotics are still active in the treatment of acinetobacter infections. Very few p-lactams are appropriate but might be useful if careful in vitro susceptibility testing has been carried out. Therapeutic choices include ticarcillin often combined with sulbactam (the latter being a Ij-lactamase inhibitor often active by itself against acinetobacter;30 ceftazidime and, most often, imipenem which is by far the most active drug in acinetobacter infection. Aminoglycosides may be used in the treatment of acinetobacter infections when combined with effective p-lactams such as ticarcillin, ceftazidime or imipenem. Combination therapy is always recommended, associating a p-lactam with an aminoglycoside, or a fluoroquinolone, or rifampicin after in vitro susceptibility testing of antibiotics and of their combinations. Control

of acinetobacter

outbreaks

in ICU

The profile of acinetobacter outbreaks fits with that of other multiplyantibiotic-resistant Gram-negative bacilli (MRGN). It has been emphasized’ that one of the major side-effects of broad-spectrum antibiotic usage is the emergence of MRGN, and among these Acinetobacter spp., which are now endemic in most hospital environments and may cause periodical outbreaks in the ICU. The reservoirs of acinetobacter are known as well as the routes of their spread. Preventive measures include isolation precautions, hand disinfection, efficient sterilization and disinfection of instruments to reduce the risk of cross-infection between patients.32 Control and surveillance programmes of prevalent organisms in ICUs is important to provide information of predominant nosocomial agents in various wards of the hospital. Prophylactic treatment with selective digestive decontamination (SDD) in ICU patients may prevent colonization of the oropharynx and of the gut by acinetobacter, the latter being one of the major colonization sites of this species in hospital patients as recently demonstrated.‘* A clear improvement in colonization and infection rates has been obtained for most MRGN33 by using SDD including most often polymyxin E, tobramycin and amphotericin (topical chemoprophylaxis). However, SDD-induced selection of resistant Gram-negative bacilli is a major concern and controversies persist as to the efficacy of SDD. Finally,

Outbreaks

of Acinetobacter

spp.

451

other means of reducing the risk of emergence of MRGN include restriction and control of the use of broad-spectrum p-lactams and surveillance of resistance in predominant bacteria. Many thanks to Marie-Lame our studies on Acinetobacter assistance.

Joly-Guillou who has contributed spp, and to Marie-Jeanne Julliard

for more than for her efficient

15 years to secretarial

References I.

2.

3. 4.

5.

6. 7. 8.

9. 10. 11. 12.

13.

14. 15. 16.

17. 18.

Bouvet PJM, Grimont PAD. Taxonomy of the genus Acinetobacter with the recognition of Acinetobacter baumannii sp. nov., Acinetobacter haemolyticus sp. nov., Acinetobacter johnsonii sp. nov., and Acinetobacter junii sp. nov. and emended descriptions of Acinetobacter calcoaceticus and Acinetobacter lwoffii. IntJ Syst Bacterial 1986; 36: 228-240. Cefai C. Richards I. Gould FK. McPeake P. An outbreak of Acinetobacter resniratorv tract infection, res&ing from incomplete disinfection of ventilatory equipment: r Hos$ Infect 1990; 5: 117-182. Gerner-Smidt P. Ribotyping of the Acinetobacter calcoaceticus-Acinetobacter baumannii comnlex. 7 Clin Microbial 1992: 30: 2680-2685. JolytGuiliou ML, DecrC D, Wolff M, Bergogne-BerCzin E. Acinetobacter spp: clinical epidemiology in 89 intensive care units. A retrospective study in France during 1991. 2nd International Conference on the Prevention of Infection (CIPI), Nice (France), +5 May 1992; Abstract CJl. Struelens MJ, Carlier E, Maes N, Serruys E, Quint WGV, van Belkum A. Nosocomial colonization and infection with multiresistant Acinetobacter baumannii: outbreak delineation using DNA macrorestriction analysis and PCR-fingerprinting. J Hosp Infect 1993; 25: 15-32. Baumann A. Isolation of Acinetobacter from soil and water. J Bacterial 1968; 96: 39-42. Bergogne-Berezin E, Joly-Guillou ML, Vieu JF. Epidemiology of nosocomial infections due to Acinetobacter calcoaceticus. J Hosp Infect 1987; 10: 105-113. Juni E. Genus III. Acinetobacter Brisou and PrCvot 1954, 727AL. In: Krieg NR, Hold JG, Eds. Bergey’s Manual of Systematic Bacteriology, Vol. 1. Baltimore: The Williams & Wilkins Co. 1984; 303-307. Ramphal R, Kluge RM. Acinetobacter calcoaceticus variety anitratus: an increasing nosocomial problem. Am J Med Sciences 1979; 277: 57-66. Noble WC, Hope YM, Midgley G et al. Toewebs as a source of Gram-negative bacilli. J Hosp Infect 1986; 8: 248-256. Taplin D, Rebel1 G, Zaias N. The human skin as a source of Mima-Herellea infections. J Am Med Assoc 1963; 186: 952-955. Timsit JF, Garrait V, Misset B, Goldstein FW, Renaud B, Carlet J. The digestive tract is a major site for Acinetobacter baumannii colonization in intensive care unit patients. J Infect Dis 1993; 168: 1336-1337. French GL, Casewell MW, Roncoroni AJ, Knoght S, Phillips I. A hospital outbreak of antibiotic-resistant Acinetobacter anitratus: epidemiology and control. J Hasp Infect 1980; 1: 125-131. Schloesser RL, Laufkoetter EA, Lehners T, Mietems C. An outbreak of Acinetobacter calcoaceticus infection in neonatal care unit. Infection 1990; 18: 230-233. Clew RH, Moellering RC, Kunz LJ. Infections with Acinetobacter calcoaceticus (Herellea vaginicola). Clinical and laboratory studies. Medicine 1977; 56: 79-87. Vila J, Almela M, Jimenez de Anta MT. Laboratory investigation of a hospital outbreak caused by two different multiresistant Acinetobacter calcoaceticus subsp. anitratus strains. J Clin Microbial 1989; 27: 1086-1089. Bouvet PJM, Grimont PAD. Identification and biotyping of clinical isolates of Acinetobacter. Ann Instit PasteurJMicrobiol 1987; 138: 569-578. Dijkshoorn L, Aucken H, Gerner-Smidt P, Kaufmann ME, Ursing J, Pitt TL. Correlation of typing methods for Acinetobacter isolates from hospital outbreaks. J Clin Microbial 1993; 31: 702-705.

452 19. 20.

21.

22.

23. 24. 25. 26. 27.

28. 29. 30. 31. 32. 33.

E. Bergogne-B&kin Gerner-Smidt P, Tjernberg I, Ursing J. Reliability of phenotypic tests for identification of Acinetobacter species. J Clin Microbial 1991; 29: 277-282. Traub WH, Spohr M. Antimicrobial drug susceptibility of clinical isolates of Acinetobacter species (A. baumannii, A. haemolyticus, genospecies 3, and genospecies 6). Antimicrob Agents Chemother 1989; 33: 1617-1619. Joly-Guillou ML, VallCe E, Bergogne-B&&in E, Philippon A. Distribution of betalactamases and phenotype analysis in clinical strains of Acinetobacter calcoaceticus. J Antimicrob Chemother 1988; 22: 597-604. Lambert T, Gerbaud G, Bouvet P, Vieu JF, Courvalin P. Dissemination of amikacin resistance gene aphA6 in Acinetobacter spp. Antimicrob Agents Chemother 1990; 34: 1244-l 248. Traub WH. Acinetobacter baumannii serotyping for delineation of outbreaks of nosocomial cross-infection. J Clin Microbial 1989; 27: 2713-2716. Vieu JF, Minck R, Bergogne-Bertzin E. Bacteriophages et lysotypie d’Acinetobacter. Ann Microbial 1979; 130A: 405-406. Andrews HJ. Acinetobacter bacteriocin typing. J Ho@ Infect 1986; 7: 169-175. Vieu JF, Bergogne-BCrCzin E, Joly ML, Berthelot G, Fichelle A. Epidemiologic d’dcinetobacter calcoaceticus. Nouv Press Med 1980; 9: 3551-3552. Alexander M, Rahman M, Taylor M, Noble W. A study of the value of electrophoretic and other techniques for typing Acinetobacter calcoaceticus. J Hasp Infect 1988; 12: 273-287. Seifert H, Schulze A, Baginski R, Pulverer G. Comparison of four different methods for epidemiologic typing of Acinetobacter baumannii. J Clin Microbial 1994; 32: 1816-I 819. Bergogne-BCrtzin E, Joly-Guillou ML. An underestimated nosocomial pathogen, Acinetobacter calcoacetzcus. J Antimicrob Chemother 1985; 16: 535-538. Seifert H, Baginski R, Schulze A, Pulverer G. Antimicrobial susceptibility of Acinetobacter species. Antimicrob Agents Chemother 1993; 37: 750-753. Urban C, Go E, Mariano N et al. Effect of sulbactam on infections caused by imipenemresistant Acinetobacter calcoaceticus biotype anitratus. J Infect Dis 1993; 67: 448451. Mehtar S. Hospital Infection Control: Setting Up a Cost-Effective Programme. New York: Oxford University Press Inc., 1992. Vandenbroucke-Grauls CM JE, Vandenbroucke JP. Effect of selective decontamination of the digestive tract on respiratory tract infections and mortality in the intensive care unit. Lancet 1991; 338: 859-862.