- Email: [email protected]
Typing of Acinetobacter baumannii isolated from hospital-acquired respiratory infections in a tertiary care centre in Southern India
Journal of Hospital Infection (2001) 47: 159–162 doi:10.1053/jhin.2000.0906, available online at http://www.idealibrary.com on
SHORT REPORT
Typing of Acinetobacter baumannii isolated from hospital-acquired respiratory infections in a tertiary care centre in Southern India E. Mathai, M. E. Kaufmann*,V. S. Richard†, G. John‡, K. N. Brahmadathan Departments of Clinical Microbiology, †Hospital Infection Control and ‡ Medicine, Christian Medical College and Hospital,Vellore,Tamil Nadu, 632004 India, and *Laboratory of Hospital Infection, Central Public Health Laboratories, 61 Colindale Avenue, London, UK
Summary: In an attempt to define the epidemiology of Acinetobacter baumannii infection, 27 isolates, obtained from hospital-acquired respiratory infections, were typed using random amplified polymorphic DNA (RAPD) profile and antimicrobial susceptibility patterns. Ten different patterns were obtained with ERIC2 primer: 14 isolates had a similar profile representing a single strain. Within RAPD types, isolates could be further classified based on their antibiogram; however, strains of different types had similar antibiograms. This study showed that many different genetic types of A. baumannii are prevalent in our hospital. While antibiograms alone are not sufficiently discriminatory, RAPD typing helps in identifying outbreaks and in assessing infection control procedures within a hospital. © 2001 The Hospital Infection Society
Keywords: Acinetobacter baumannii; hospital-acquired pneumonia; RAPD typing; outbreak.
Introduction In recent years, Acinetobacter baumannii has emerged as an important cause of hospitalacquired pneumonia, especially in patients with severe underlying diseases, or on ventilatory support in intensive care units.1–6 Due to its intrinsic resistance to a variety of antibiotics, severe infections due to this organism are associated with high mortality rates.1–3 This feature, together with the ability of the organism to survive long-term in the hospital environment and to be transmitted, via human or inanimate sources in outbreaks makes it a matter of serious concern in hospital infection control practice.1–5 Once established in the hospital Received 13 July 2000; revised manuscript accepted 1 November 2000. Author for correspondence: Professor E. Mathai, Dept. of Clinical Microbiology, Christian Medical College and Hospital, Vellore, Tamil Nadu, India 632004.
0195-6701/01/020159;04 $35.00
patient environment, many strains prove very difficult to eradicate by conventional infection control procedures.5 The genus Acinetobacter comprises of at least 19 genomic species as defined by DNA–DNA hybridization.7 The most prevalent in hospitalacquired infection is the so-called Acinetobacter calcoaceticus-baumannii (ACB) complex containing the closely related glucose-oxidizing named species, A. calcoaceticus, A. baumannii, and the unnamed DNA groups 3 and 13.7 Of these, A. baumannii is the most frequent cause of nosocomial infection.1,7 Unfortunately no single phenotypic test or short set of tests provides unequivocal identification and differentiation of the genomic species. Of the many methods described for typing ACB complex, the best discrimination and reproducibility are obtained with DNA fingerprinting methods such as pulsed-field gel electrophoresis (PFGE) of © 2001 The Hospital Infection Society
160
restriction endonuclease digests of chromosomal DNA and random amplified polymorphic DNA (RAPD) profiles using repetitive element PCR.5,6,8 During October and November 1998, several isolates of A. baumannii were obtained from patients with nosocomial respiratory infections. These isolates were further tested in an attempt to determine the epidemiology of A. baumannii infection in our hospital. Methods Sputum and suction tips from patients, with suspected nosocomial pneumonia, were routinely cultured on sheep blood agar and MacConkey’s agar and incubated overnight at 37⬚C. Glucose oxidizing, oxidase-negative, non-motile Gram-negative bacilli, growing on MacConkey’s agar were identified as belonging to A. calcoaceticus baumannii complex.7 The isolates included in the study grew at 44⬚C. During the study period, 48 distinct clinical isolates of A. baumannii (14 from medical ICU, 13 from neurology ICU, six from surgical ICU, five from surgical wards, four from a paediatric ward and six from other parts of the hospital) were cultured. Of these, 27 isolates (eight each from medical and neurology ICU, three each from the surgical ICU and paediatrics, two each from medical and surgical wards and one isolate from nursery) were further typed at the Central Public Health Laboratory, London. The minimum inhibitory concentrations (MIC) for amikacin, ceftazidime, ciprofloxacin, gentamicin, imipenem, meropenem, piperacillin and piperacillin/tazobactam combination were determined by dilution in Isosensitest (Oxoid) agar supplemented with 2% lysed horse blood. The isolates were grown overnight in nutrient broth and approximately 105 cfu/spot was applied to the antibiotic plates. Two isolates with raised MIC for imipenem were tested by PCR for the Imp gene.9 The genetic relatedness of isolates was determined by RAPD analysis using ERIC2 primers following an SOP provided by the Laboratory of Hospital Infection.
E. Mathai et al.
profile representing a single strain, designated A, three isolates had strain pattern F, two each had B and E, and one each had patterns C, D, G, H, I and J. Three isolates (one each of A, E and F) differed from the major type by a single band and were therefore considered variants. Thirteen isolates of strain A were from adults with ventilator-associated pneumonia (VAP) on intensive care units. Two groups of these isolates were distinguished by the antibiogram (Table I). The first group of 11 isolates was resistant to all antibiotics tested except carbapenems and the three isolates of the second group were also susceptible to aminoglycosides. These two groups were isolated from distinct areas in the hospital. The MIC of imipenem was generally higher for strain A (1–4 mg/L), compared to other strain types (less than : 0.25 mg/L). Six of the initial strain A isolates were obtained during a period of six days from the medical ICU. Infection control measures were strictly implemented during this period and only one more isolate was identified from this area, a few days later. However, strain A was subsequently isolated from other parts of the hospital. Infection with this strain was fatal in seven of the 13 (54%) patients. Strain B isolates were obtained four days apart from babies on ventilators. Strains of types E and F were obtained from surgical ICU and postoperative patients. Mortality was high 3/5 (60%) in this group also, though these isolates were susceptible to many antibiotics. All isolates were susceptible to carbapenem antibiotics. Two isolates with raised imipenem MICs belonging to RAPD type A did not have Imp gene. Only 11 of the 27 (41%) isolates were susceptible to both gentamicin and amikacin. Nine (33%) isolates were susceptible to ceftazidime and five (19%) to ciprofloxacin. Only four (16%) of the 25 strains resistant to piperacillin were susceptible to piperacillin–tazobactam combination. Repeat isolates obtained from two patients in the medical ICU belonged to the same group A, however, a third patient in neurology ICU harboured isolates of types C, A and I within a few days of each other.
Results The 27 isolates were obtained from 23 patients. Apart from four sputum isolates, all the others were from suction tips used on patients on ventilators. Ten different RAPD patterns were obtained with ERIC2 primers; 14 isolates had a similar
Discussion The RAPD profiles of the isolates indicate the existence of different clones of A. baumannii in distinct areas of our hospital. Data from other centres also show that different clones of A. baumannii can
Typing of A. baumannii
161
Table 1
Antibiogram and distribution of different ERIC2 RAPD types
Type
Number
Antibiotics susceptible
Ward isolated
Ventilated
Fatal
A
11
I, M
11
3;3*
3
A, G, I, M
2
1
C
1 1 1
D E E’
1 1 1
F
1
F’ G
1 1 1
H I
1 1
J
1
I, M A, I, M A, G, Cip, CTZ, I, M, P/T CTZ, I, M I, M A, G, Cip, CTZ, I, M, P/T A, G, CTZ, I, M, P, P/T A, G, CTZ, I, M A, G, CTZ, I, M A, G, Cip, CTZ, I, M, P, P/T G, I, M, A, G, Cip, CTZ, I, M, P/T A, G, Cip, CTZ, I, M, P/T
MICU (7) NICUB (3) Nursery (1) NICUA (2) C (1) Q3W Q3W NICU
B
1 1 1
NICU 1 SICU 1 KN (post op)
1
SICU
1
2
H (Post op) SICU I
1
MICU NICU
1 1
Q3W
1
A, amikacin; G, gentamicin; Cip, ciprofloxacin; CTZ, ceftazidime; I, imipenem; M, meropenem; P, piperacillin; P/T, piperacilin/tazobactam; MICU, medical ICU; NICU, neurology ICU; SICU, surgical ICU; Q3W, paediatric ward. *Discharged on request when seriously ill, presumed dead.
be prevalent in the same hospital at the same time or over a period of time.2,5,6,8 In a recent study one multi-resistant strain of A. baumannii predominated in the ICU of a hospital with well developed infection control programme as against multiple clones in another hospital with a poor infection control system.10 Our data therefore probably indicate the need for tighter infection control practices. The potential of this organism to cause outbreaks is well known.1,5 During the study period, there was one major outbreak due to a multidrug-resistant A. baumannii strain, A, in the ICU and the strain probably spread from one ICU to another. How this happened is not clear, since according to hospital protocols, there is no exchange of equipment or personnel between ICUs. Strict implementation of conventional infection control measures helped to curb this outbreak. Complete eradication of this organism from hospital environment is extremely difficult,5 as we observed. The data also showed cross infections due to specific strain types occurring in various parts of the hospital. Therefore, in addition to determining
the epidemiology, typing also helps in assessing the efficiency of infection control measures. Different antibiograms were observed within one RAPD type, although, some strains belonging to different genetic types had similar antibiograms. As has been reported,3,8 antibiograms alone are not useful in identifying outbreaks, unless used in conjunction with RAPD profile, and thereby improve discriminatory power. While some strains were susceptible to many different antibiotics, others were multi-resistant. Overall, ceftazidime and piperacillin susceptibility rates were lower than that reported from elsewhere.11 Prevalence of such drug-resistant strains poses a problem with therapy, especially in resource poor countries like India. Importance of strict infection control measures in such places cannot be over emphasized.12 Acknowledgement The authors wish to thank Dr T. L. Pitt, Deputy Director, Laboratory of Hospital Infection, Central
162
Public Health Laboratories, 61 Colindale Avenue, London for RAPD typing and MIC testing of the strains included in the study. References 1. Bergogne Berezin E, Towner KT. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical and epidemiological features. Clin Microbiol Rev 1996; 9: 148–165. 2. Rello J, Torres A. Microbial causes of ventilator-associated pneumonia. Semin Respir Infect 1996; 11: 24–31. 3. Husni RN, Goldstein LS, Arroliga AC, Hall GS, Fatica C, Stoller JK, Gordon SM. Risk factors for an outbreak of multi drug resistant Acinetobacter nosocomial pneumonia among intubated patients. Chest 1999; 115: 1378–1382. 4. Dy ME, Nord JA, LaBombardi VJ, Kislak JW. The emergence of resistant strains of Acinetobacter baumannii: clinical and infection control implications. Infect Control Hosp Epidemiol 1999; 20: 565–567. 5. Riley TV, Webb SA, Cadwallader H, Briggs BD, Christiansen L, Bowman RA. Out break of gentamicin resistant Acinetobacter baumannii in an intensive care unit: clinical, epidemiological and microbiological features. Pathology 1996; 28: 359–363. 6. Christie C, Mazon D, Hierholzer W Jr, Patterson JE. Molecular heterogeneity of Acinetobacter baumannii isolates during seasonal increase in prevalence. Infect Control Hosp Epidemiol 1995; 16: 590–594.
E. Mathai et al.
07. The non-fermentative Gram Negative Bacteria. In: Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC Jr, Eds. Color Atlas and text book of diagnostic microbiology 5th edn Philadelphia, USA Lippincott, 1997; 253–320. 08. Sheehan C, Boissel L, Lynch M, Cryan B, Fanning S. DNA amplification finger printing (DAF) applied to investigation of Acinetobacter baumannii isolates from Intensive Care Unit patients. Ir J Med Sci 1996; 165: 44–48. 09. Senda K, Arakawa Y, Ichiyama S et al. PCR detection of metallo beta lactamase gene (blaimp) in gramnegative rods resistant to broad-spectrum lactams. J Clin Microbiol 1996; 34: 2909–2913. 10. Webster CA, Towner KJ, Saunders GL, CreweBrown HH, Humphreys H. Molecular and anti biogram relationships of Acinetobacter isolates from two contrasting hospitals in the United Kingdom and South Africa. Eur J Clin Microbiol Infect Dis 1999; 18: 595–598. 11. The Venezuelan antimicrobial resistance study group, Pfaller MA, Jones RN, Doern GV. Multicenter evaluation of the antimicrobial activity to six broad spectrum  lactams in Venezuela: comparison of data from 1997 and 1998 using the E test method. Diagn Microbiol Infect Dis 1999; 35: 153–158. 12. Mathai E, Richard V, John G, Brahmadathan KN. Typing of Acinetobacter baumannii isolated from respiratory infection. Clin Microbiol Infect 2000; 6: 121.
SHORT REPORT
Typing of Acinetobacter baumannii isolated from hospital-acquired respiratory infections in a tertiary care centre in Southern India E. Mathai, M. E. Kaufmann*,V. S. Richard†, G. John‡, K. N. Brahmadathan Departments of Clinical Microbiology, †Hospital Infection Control and ‡ Medicine, Christian Medical College and Hospital,Vellore,Tamil Nadu, 632004 India, and *Laboratory of Hospital Infection, Central Public Health Laboratories, 61 Colindale Avenue, London, UK
Summary: In an attempt to define the epidemiology of Acinetobacter baumannii infection, 27 isolates, obtained from hospital-acquired respiratory infections, were typed using random amplified polymorphic DNA (RAPD) profile and antimicrobial susceptibility patterns. Ten different patterns were obtained with ERIC2 primer: 14 isolates had a similar profile representing a single strain. Within RAPD types, isolates could be further classified based on their antibiogram; however, strains of different types had similar antibiograms. This study showed that many different genetic types of A. baumannii are prevalent in our hospital. While antibiograms alone are not sufficiently discriminatory, RAPD typing helps in identifying outbreaks and in assessing infection control procedures within a hospital. © 2001 The Hospital Infection Society
Keywords: Acinetobacter baumannii; hospital-acquired pneumonia; RAPD typing; outbreak.
Introduction In recent years, Acinetobacter baumannii has emerged as an important cause of hospitalacquired pneumonia, especially in patients with severe underlying diseases, or on ventilatory support in intensive care units.1–6 Due to its intrinsic resistance to a variety of antibiotics, severe infections due to this organism are associated with high mortality rates.1–3 This feature, together with the ability of the organism to survive long-term in the hospital environment and to be transmitted, via human or inanimate sources in outbreaks makes it a matter of serious concern in hospital infection control practice.1–5 Once established in the hospital Received 13 July 2000; revised manuscript accepted 1 November 2000. Author for correspondence: Professor E. Mathai, Dept. of Clinical Microbiology, Christian Medical College and Hospital, Vellore, Tamil Nadu, India 632004.
0195-6701/01/020159;04 $35.00
patient environment, many strains prove very difficult to eradicate by conventional infection control procedures.5 The genus Acinetobacter comprises of at least 19 genomic species as defined by DNA–DNA hybridization.7 The most prevalent in hospitalacquired infection is the so-called Acinetobacter calcoaceticus-baumannii (ACB) complex containing the closely related glucose-oxidizing named species, A. calcoaceticus, A. baumannii, and the unnamed DNA groups 3 and 13.7 Of these, A. baumannii is the most frequent cause of nosocomial infection.1,7 Unfortunately no single phenotypic test or short set of tests provides unequivocal identification and differentiation of the genomic species. Of the many methods described for typing ACB complex, the best discrimination and reproducibility are obtained with DNA fingerprinting methods such as pulsed-field gel electrophoresis (PFGE) of © 2001 The Hospital Infection Society
160
restriction endonuclease digests of chromosomal DNA and random amplified polymorphic DNA (RAPD) profiles using repetitive element PCR.5,6,8 During October and November 1998, several isolates of A. baumannii were obtained from patients with nosocomial respiratory infections. These isolates were further tested in an attempt to determine the epidemiology of A. baumannii infection in our hospital. Methods Sputum and suction tips from patients, with suspected nosocomial pneumonia, were routinely cultured on sheep blood agar and MacConkey’s agar and incubated overnight at 37⬚C. Glucose oxidizing, oxidase-negative, non-motile Gram-negative bacilli, growing on MacConkey’s agar were identified as belonging to A. calcoaceticus baumannii complex.7 The isolates included in the study grew at 44⬚C. During the study period, 48 distinct clinical isolates of A. baumannii (14 from medical ICU, 13 from neurology ICU, six from surgical ICU, five from surgical wards, four from a paediatric ward and six from other parts of the hospital) were cultured. Of these, 27 isolates (eight each from medical and neurology ICU, three each from the surgical ICU and paediatrics, two each from medical and surgical wards and one isolate from nursery) were further typed at the Central Public Health Laboratory, London. The minimum inhibitory concentrations (MIC) for amikacin, ceftazidime, ciprofloxacin, gentamicin, imipenem, meropenem, piperacillin and piperacillin/tazobactam combination were determined by dilution in Isosensitest (Oxoid) agar supplemented with 2% lysed horse blood. The isolates were grown overnight in nutrient broth and approximately 105 cfu/spot was applied to the antibiotic plates. Two isolates with raised MIC for imipenem were tested by PCR for the Imp gene.9 The genetic relatedness of isolates was determined by RAPD analysis using ERIC2 primers following an SOP provided by the Laboratory of Hospital Infection.
E. Mathai et al.
profile representing a single strain, designated A, three isolates had strain pattern F, two each had B and E, and one each had patterns C, D, G, H, I and J. Three isolates (one each of A, E and F) differed from the major type by a single band and were therefore considered variants. Thirteen isolates of strain A were from adults with ventilator-associated pneumonia (VAP) on intensive care units. Two groups of these isolates were distinguished by the antibiogram (Table I). The first group of 11 isolates was resistant to all antibiotics tested except carbapenems and the three isolates of the second group were also susceptible to aminoglycosides. These two groups were isolated from distinct areas in the hospital. The MIC of imipenem was generally higher for strain A (1–4 mg/L), compared to other strain types (less than : 0.25 mg/L). Six of the initial strain A isolates were obtained during a period of six days from the medical ICU. Infection control measures were strictly implemented during this period and only one more isolate was identified from this area, a few days later. However, strain A was subsequently isolated from other parts of the hospital. Infection with this strain was fatal in seven of the 13 (54%) patients. Strain B isolates were obtained four days apart from babies on ventilators. Strains of types E and F were obtained from surgical ICU and postoperative patients. Mortality was high 3/5 (60%) in this group also, though these isolates were susceptible to many antibiotics. All isolates were susceptible to carbapenem antibiotics. Two isolates with raised imipenem MICs belonging to RAPD type A did not have Imp gene. Only 11 of the 27 (41%) isolates were susceptible to both gentamicin and amikacin. Nine (33%) isolates were susceptible to ceftazidime and five (19%) to ciprofloxacin. Only four (16%) of the 25 strains resistant to piperacillin were susceptible to piperacillin–tazobactam combination. Repeat isolates obtained from two patients in the medical ICU belonged to the same group A, however, a third patient in neurology ICU harboured isolates of types C, A and I within a few days of each other.
Results The 27 isolates were obtained from 23 patients. Apart from four sputum isolates, all the others were from suction tips used on patients on ventilators. Ten different RAPD patterns were obtained with ERIC2 primers; 14 isolates had a similar
Discussion The RAPD profiles of the isolates indicate the existence of different clones of A. baumannii in distinct areas of our hospital. Data from other centres also show that different clones of A. baumannii can
Typing of A. baumannii
161
Table 1
Antibiogram and distribution of different ERIC2 RAPD types
Type
Number
Antibiotics susceptible
Ward isolated
Ventilated
Fatal
A
11
I, M
11
3;3*
3
A, G, I, M
2
1
C
1 1 1
D E E’
1 1 1
F
1
F’ G
1 1 1
H I
1 1
J
1
I, M A, I, M A, G, Cip, CTZ, I, M, P/T CTZ, I, M I, M A, G, Cip, CTZ, I, M, P/T A, G, CTZ, I, M, P, P/T A, G, CTZ, I, M A, G, CTZ, I, M A, G, Cip, CTZ, I, M, P, P/T G, I, M, A, G, Cip, CTZ, I, M, P/T A, G, Cip, CTZ, I, M, P/T
MICU (7) NICUB (3) Nursery (1) NICUA (2) C (1) Q3W Q3W NICU
B
1 1 1
NICU 1 SICU 1 KN (post op)
1
SICU
1
2
H (Post op) SICU I
1
MICU NICU
1 1
Q3W
1
A, amikacin; G, gentamicin; Cip, ciprofloxacin; CTZ, ceftazidime; I, imipenem; M, meropenem; P, piperacillin; P/T, piperacilin/tazobactam; MICU, medical ICU; NICU, neurology ICU; SICU, surgical ICU; Q3W, paediatric ward. *Discharged on request when seriously ill, presumed dead.
be prevalent in the same hospital at the same time or over a period of time.2,5,6,8 In a recent study one multi-resistant strain of A. baumannii predominated in the ICU of a hospital with well developed infection control programme as against multiple clones in another hospital with a poor infection control system.10 Our data therefore probably indicate the need for tighter infection control practices. The potential of this organism to cause outbreaks is well known.1,5 During the study period, there was one major outbreak due to a multidrug-resistant A. baumannii strain, A, in the ICU and the strain probably spread from one ICU to another. How this happened is not clear, since according to hospital protocols, there is no exchange of equipment or personnel between ICUs. Strict implementation of conventional infection control measures helped to curb this outbreak. Complete eradication of this organism from hospital environment is extremely difficult,5 as we observed. The data also showed cross infections due to specific strain types occurring in various parts of the hospital. Therefore, in addition to determining
the epidemiology, typing also helps in assessing the efficiency of infection control measures. Different antibiograms were observed within one RAPD type, although, some strains belonging to different genetic types had similar antibiograms. As has been reported,3,8 antibiograms alone are not useful in identifying outbreaks, unless used in conjunction with RAPD profile, and thereby improve discriminatory power. While some strains were susceptible to many different antibiotics, others were multi-resistant. Overall, ceftazidime and piperacillin susceptibility rates were lower than that reported from elsewhere.11 Prevalence of such drug-resistant strains poses a problem with therapy, especially in resource poor countries like India. Importance of strict infection control measures in such places cannot be over emphasized.12 Acknowledgement The authors wish to thank Dr T. L. Pitt, Deputy Director, Laboratory of Hospital Infection, Central
162
Public Health Laboratories, 61 Colindale Avenue, London for RAPD typing and MIC testing of the strains included in the study. References 1. Bergogne Berezin E, Towner KT. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical and epidemiological features. Clin Microbiol Rev 1996; 9: 148–165. 2. Rello J, Torres A. Microbial causes of ventilator-associated pneumonia. Semin Respir Infect 1996; 11: 24–31. 3. Husni RN, Goldstein LS, Arroliga AC, Hall GS, Fatica C, Stoller JK, Gordon SM. Risk factors for an outbreak of multi drug resistant Acinetobacter nosocomial pneumonia among intubated patients. Chest 1999; 115: 1378–1382. 4. Dy ME, Nord JA, LaBombardi VJ, Kislak JW. The emergence of resistant strains of Acinetobacter baumannii: clinical and infection control implications. Infect Control Hosp Epidemiol 1999; 20: 565–567. 5. Riley TV, Webb SA, Cadwallader H, Briggs BD, Christiansen L, Bowman RA. Out break of gentamicin resistant Acinetobacter baumannii in an intensive care unit: clinical, epidemiological and microbiological features. Pathology 1996; 28: 359–363. 6. Christie C, Mazon D, Hierholzer W Jr, Patterson JE. Molecular heterogeneity of Acinetobacter baumannii isolates during seasonal increase in prevalence. Infect Control Hosp Epidemiol 1995; 16: 590–594.
E. Mathai et al.
07. The non-fermentative Gram Negative Bacteria. In: Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC Jr, Eds. Color Atlas and text book of diagnostic microbiology 5th edn Philadelphia, USA Lippincott, 1997; 253–320. 08. Sheehan C, Boissel L, Lynch M, Cryan B, Fanning S. DNA amplification finger printing (DAF) applied to investigation of Acinetobacter baumannii isolates from Intensive Care Unit patients. Ir J Med Sci 1996; 165: 44–48. 09. Senda K, Arakawa Y, Ichiyama S et al. PCR detection of metallo beta lactamase gene (blaimp) in gramnegative rods resistant to broad-spectrum lactams. J Clin Microbiol 1996; 34: 2909–2913. 10. Webster CA, Towner KJ, Saunders GL, CreweBrown HH, Humphreys H. Molecular and anti biogram relationships of Acinetobacter isolates from two contrasting hospitals in the United Kingdom and South Africa. Eur J Clin Microbiol Infect Dis 1999; 18: 595–598. 11. The Venezuelan antimicrobial resistance study group, Pfaller MA, Jones RN, Doern GV. Multicenter evaluation of the antimicrobial activity to six broad spectrum  lactams in Venezuela: comparison of data from 1997 and 1998 using the E test method. Diagn Microbiol Infect Dis 1999; 35: 153–158. 12. Mathai E, Richard V, John G, Brahmadathan KN. Typing of Acinetobacter baumannii isolated from respiratory infection. Clin Microbiol Infect 2000; 6: 121.