Eur J Pediatr DOI 10.1007/s00431-014-2365-8
ORIGINAL ARTICLE
Successful management of an outbreak due to carbapenem-resistant Acinetobacter baumannii in a neonatal intensive care unit Olga Tsiatsiou & Εlias Iosifidis & Aspasia Katragkou & Vasiliki Dimou & Kosmas Sarafidis & Theodoros Karampatakis & Charalampos Antachopoulos & Anagnostina Orfanou & Athanasios Tsakris & Vasiliki Drossou-Agakidou & Emmanuel Roilides
Received: 23 March 2014 / Revised: 1 June 2014 / Accepted: 16 June 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract The investigation and successful management of a monoclonal Acinetobacter baumannii outbreak in a neonatal intensive care unit are described. Upon the first clustered carbapenem-resistant A. baumannii (CRAB) infections, a bundle of actions were taken, including enhanced infection control, active surveillance (weekly stool samples), casecontrol study, staff education, daily audits and discontinuation of new admissions. Between September and December 2011,
eight neonates developed 10 CRAB infections (five blood, four respiratory and one eye). A total of 216 active surveillance cultures were obtained from 96 neonates (43 % had ≥2 samples). During weeks 12, 16 and 17, active surveillance detected 3, 1 and 2 new CRAB acquisitions, respectively. Prevalence of infections/colonizations decreased, and no event occurred after 20th week. A colonized neonate developed CRAB sepsis and died. All CRAB isolates harboured
Communicated by David Nadal Εlias Iosifidis and Aspasia Katragkou equally contributed to the study. This work was presented in part as a poster to the 30th Annual Meeting of the European Society for Pediatric Infectious Diseases in Thessaloniki Greece, May 8–12, 2012. O. Tsiatsiou : Ε. Iosifidis : A. Katragkou : C. Antachopoulos : E. Roilides (*) Infectious Diseases Unit, 3rd Department of Pediatrics, Aristotle University School of Medicine, Hippokration General Hospital, Konstantinoupoleos 49, 546 42 Thessaloniki, Greece e-mail: [email protected] O. Tsiatsiou e-mail: [email protected] Ε. Iosifidis e-mail: [email protected] A. Katragkou e-mail: [email protected] C. Antachopoulos e-mail: [email protected]
V. Dimou e-mail: [email protected] T. Karampatakis e-mail: [email protected] A. Orfanou e-mail: [email protected] K. Sarafidis : V. Drossou-Agakidou 1st Department of Neonatology and Neonatal Intensive Care Unit, Aristotle University School of Medicine, Hippokration General Hospital, Thessaloniki, Greece K. Sarafidis e-mail: [email protected] V. Drossou-Agakidou e-mail: [email protected]
A. Katragkou Division of Infectious Diseases, Weill Cornell Medical Center, Cornell University, New York, NY, USA V. Dimou : T. Karampatakis : A. Orfanou Microbiology Department, Hippokration General Hospital, Thessaloniki, Greece
A. Tsakris Microbiology Department, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece e-mail: [email protected]
Eur J Pediatr
bla OXA-58 and the intrinsic chromosomal bla OXA-51 carbapenemase genes. Conclusion: Active surveillance and enhanced infection control measures effectively contained spread of CRAB clone in the neonatal intensive care unit. Keywords Acinetobacter baumannii . Neonate . Outbreak . Carbapenem . Carbapenemase Abbreviations CRAB ID MIC NICU PFGE PICU
Carbapenem-resistant Acinetobacter baumannii Infectious disease Minimum inhibitory concentration Neonatal intensive care unit Pulsed-field gel electrophoresis Pediatric intensive care unit
Hippokration Hospital of Thessaloniki. This is a 920-bed tertiary-care general hospital treating both adult and paediatric patients. Apart from the NICU, there is a paediatric and an adult intensive care unit. The specific level 3 NICU has 44 beds including 15 intensive care beds providing critical care to neonates born in the same or other hospitals. In 2011, the NICU consisted of six wards, but there were no single rooms for isolation. A total of 674 neonates were admitted, while the occupancy rate was 80 % and the annual mortality rate 5 %. In this hospital, there are three infection control nurses and two infectious disease (ID) physicians as well as three ID fellows. Since 2011, there is active surveillance of four hospital-associated infections (bloodstream infections, urinary tract infections, lower respiratory tract infections and surgical site infections) caused by carbapenem-resistant bacteria [18]. Contact precautions are standard for patients with carbapenem-resistant bacteria. Study design
Introduction Acinetobacter baumannii has been increasingly recognized as an important nosocomial pathogen particularly in the intensive care settings [5]. The propensity of Acinetobacter spp. to grow at a range of different temperatures and pH values favours their dissemination in the hospital environment [5]. During the past 30 years, clones of Acinetobacter spp. have gradually accumulated resistance to penicillins, cephalosporins, quinolones and aminoglycosides while an increasing number of carbapenem-resistant Acinetobacter isolates have been reported worldwide [1, 32]. The emergence of the latter phenomenon is particularly concerning especially in neonates, which are very susceptible to infections and in which therapeutic options with antibiotics are quite limited. In the present study, an outbreak of carbapenem-resistant A. baumannii (CRAB) infections in a Greek neonatal intensive care unit (NICU) is reported. During the last months of 2011, an increased number of infections due to CRAB were noticed. The severe clinical manifestations caused by these isolates and their resistance phenotype prompted an urgent and intense investigation. Herein, we analyze the characteristics of the patients infected or colonized by CRAB and describe the methods used in order to investigate and successfully control the outbreak caused by a single clone of CRAB containing blaOXA-58 carbapenemase gene.
Patients and methods Clinical setting The outbreak occurred at the 1st Department of Neonatology and NICU of the Aristotle University located in the
This was a prospective study including the identification and investigation of the outbreak as well as the implementation of all bundle’s elements used for the management of this outbreak. The investigation was considered to be an infection control response, and as such, it was not subjected to approval by an institutional review board or required written informed consent from patients. Identification and investigation of the outbreak Prior to September 2011, no CRAB infections had occurred in the NICU. In September 2011, the first case of CRAB bloodstream infection occurred in a neonate on day 6 of hospitalization. After the occurrence of a second case of CRAB infection, an outbreak team was formed and consisted of two infection control nurses, the head of the microbiology department and paediatric ID physicians, all of whom worked together with the neonatologists and the other NICU staff. Case patients were defined as neonates with a nosocomial infection caused by a CRAB isolate identified in a clinically significant culture. Nosocomial infection was defined according to the criteria used by the Centers for Disease Control and Prevention [12]. A CRAB isolate was defined as having a minimum inhibitory concentration (MIC) of imipenem equal to or higher than 8 mg/L according to the interpretative criteria of the Clinical and Laboratory Standards Institute. This included intermediately resistant isolates [15]. Environmental cultures from places such as monitors, tables, patient charts and mechanical ventilation equipment were obtained, and a case-control study was conducted in order to identify possible risk factors associated with CRAB infection. For this purpose, cases were matched to controls (1:2 ratio), which were randomly selected among neonates
Eur J Pediatr
receiving intensive care during the same period for at least 48 h but had no positive cultures with CRAB. Demographic and clinical data recorded included date of birth, sex, gestational age, birth weight, underlying disease, invasive procedures, antimicrobial therapies and clinical outcome. Patient location and isolate assignment were recorded in the unit daily to determine potential modes of CRAB spread. Data collection and weekly epidemiologic surveillance were initiated shortly after the outbreak was noted.
Active surveillance for CRAB colonization
Implementation of all bundle’s elements
Temporary closure of the NICU
As soon as the outbreak was identified and investigation started, a bundle of actions were implemented as follow: Apart from the formation of the outbreak management team, the case-control study and the environmental cultures, other measures included the following: intensification of infection control measures, implementation of active surveillance cultures for CRAB colonization in hospitalized neonates, optimization of antimicrobial therapy and closure of the department to new admissions. Specifically, no neonates were admitted for 12 days; then, the department was admitting only neonates born in the same hospital for an additional week.
Temporary discontinuation of accepting new admissions to the NICU was decided after the results of the first active surveillance cultures during the 12th week of this outbreak.
Intensified infection control measures
Microbiologic methods and molecular typing
Intensification of infection control measures included the following: (1) Cohorting of neonates colonized or infected with CRAB in designated areas of the NICU. The newly admitted neonates were separated from the infected or colonized ones and managed by different medical and nursing teams (staff cohorting); (2) monitoring of adherence to contact precautions applied to neonates with culture results positive for CRAB and to standard precautions for all other neonates; (3) reevaluation and improvement of environmental cleaning and decontamination as well as the sterilization process of reusable equipment and identification of potential flaws in standard NICU practices. Processes such as intravenous access, use of heparin locks, preparation of intravenously administered fluids and use of syringes were reviewed and directly observed by the infection control team; (4) availability of replaceable dispensers containing alcohol hand sanitizer in all places for immediate access such as incubators, trolleys, nurse station and walls was regularly checked; (5) monitoring and evaluation of the adherence to infection control measures by the infection control nurses at various times of the day and (6) education of all NICU staff and neonates’ families by the ID physicians and infection control nurses on the importance of strict adherence to all enhanced infection control measures.
All clinical samples regularly obtained as clinically indicated for culture from all patients in the unit were processed using standard practice and culture media. Blood cultures were processed using the BacT/ALERT automated system (bioMérieux, Hazelwood, MO). Identification and susceptibility testing of A. baumannii isolates were performed by the microbiology laboratory using the Vitek 2 automated system (bioMérieux). The MIC determination and interpretation results from the automated system were comparable to those of the Clinical and Laboratory Standards Institute [15]. Imip en em MICs we re a lso c onf irmed by E- test (bioMérieux). The active surveillance specimens were inoculated in MacConkey agar plates supplemented with imipenem (1 mg/L). After overnight incubation at 37 °C, suspected Acinetobacter spp. colonies were further examined. Identification to the species level and antimicrobial susceptibility testing were performed as described above. Genes encoding metallo-β-lactamases (blaVIM, blaIMP and bla SIM ) and oxacillinase (OXA)-type carbapenemases (blaOXA51-like, blaOXA23-like, blaOXA24-like and blaOXA58-like) were sought by PCR amplification using primers and conditions previously published [10, 34]. For the genotypic analysis of the isolates, chromosomal DNA was digested using the ApaI restriction endonuclease (Fermendas, Athens, Greece), and the generated fragments were analyzed by pulsed-field gel
Active surveillance for CRAB colonization was started on the 12th week of the outbreak and continued until the 21st week. All neonates hospitalized in the NICU were eligible. More specific, perianal or stool samples were collected weekly from all neonates until discharge. Colonization with CRAB was defined as the acquisition of this pathogen during hospitalization (at least one positive surveillance culture) but without any evidence of clinical disease.
Changes in antibiotic policy First-line empirical antibiotic therapy including ampicillin with gentamicin was not modified during this outbreak. For the treatment of late-onset sepsis, both antibiotic regimen and duration were discussed daily with ID physicians. Colistin and carbapenems were administered only after ID approval during the outbreak period.
Eur J Pediatr
electrophoresis (PFGE) in a CHEF-DR III apparatus (Bio-Rad Laboratories, Athens, Greece) at 14 °C with a running time of 21 h and pulse times ranging from 3 to 40 s [27]. Banding patterns were compared visually according to the criteria suggested by Tenover et al. [29]. Clinical outcomes Daily prevalence of neonates infected by CRAB was calculated throughout the outbreak period. After the implementation of active surveillance cultures, daily prevalence of all infected or colonized neonates by CRAB was also assessed. All cause- and infection- related mortality was monitored during the outbreak period. Statistical analysis All continuous variables were expressed as mean±SEM (standard error of mean). In case-control study, Student’s t test or Mann-Whitney U test was used for the comparison of normally distributed and non-normally distributed continuous variables, respectively. For categorical variables, the Fisher’s exact test was used in univariate analysis. A two tailed p value of
ORIGINAL ARTICLE
Successful management of an outbreak due to carbapenem-resistant Acinetobacter baumannii in a neonatal intensive care unit Olga Tsiatsiou & Εlias Iosifidis & Aspasia Katragkou & Vasiliki Dimou & Kosmas Sarafidis & Theodoros Karampatakis & Charalampos Antachopoulos & Anagnostina Orfanou & Athanasios Tsakris & Vasiliki Drossou-Agakidou & Emmanuel Roilides
Received: 23 March 2014 / Revised: 1 June 2014 / Accepted: 16 June 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract The investigation and successful management of a monoclonal Acinetobacter baumannii outbreak in a neonatal intensive care unit are described. Upon the first clustered carbapenem-resistant A. baumannii (CRAB) infections, a bundle of actions were taken, including enhanced infection control, active surveillance (weekly stool samples), casecontrol study, staff education, daily audits and discontinuation of new admissions. Between September and December 2011,
eight neonates developed 10 CRAB infections (five blood, four respiratory and one eye). A total of 216 active surveillance cultures were obtained from 96 neonates (43 % had ≥2 samples). During weeks 12, 16 and 17, active surveillance detected 3, 1 and 2 new CRAB acquisitions, respectively. Prevalence of infections/colonizations decreased, and no event occurred after 20th week. A colonized neonate developed CRAB sepsis and died. All CRAB isolates harboured
Communicated by David Nadal Εlias Iosifidis and Aspasia Katragkou equally contributed to the study. This work was presented in part as a poster to the 30th Annual Meeting of the European Society for Pediatric Infectious Diseases in Thessaloniki Greece, May 8–12, 2012. O. Tsiatsiou : Ε. Iosifidis : A. Katragkou : C. Antachopoulos : E. Roilides (*) Infectious Diseases Unit, 3rd Department of Pediatrics, Aristotle University School of Medicine, Hippokration General Hospital, Konstantinoupoleos 49, 546 42 Thessaloniki, Greece e-mail: [email protected] O. Tsiatsiou e-mail: [email protected] Ε. Iosifidis e-mail: [email protected] A. Katragkou e-mail: [email protected] C. Antachopoulos e-mail: [email protected]
V. Dimou e-mail: [email protected] T. Karampatakis e-mail: [email protected] A. Orfanou e-mail: [email protected] K. Sarafidis : V. Drossou-Agakidou 1st Department of Neonatology and Neonatal Intensive Care Unit, Aristotle University School of Medicine, Hippokration General Hospital, Thessaloniki, Greece K. Sarafidis e-mail: [email protected] V. Drossou-Agakidou e-mail: [email protected]
A. Katragkou Division of Infectious Diseases, Weill Cornell Medical Center, Cornell University, New York, NY, USA V. Dimou : T. Karampatakis : A. Orfanou Microbiology Department, Hippokration General Hospital, Thessaloniki, Greece
A. Tsakris Microbiology Department, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece e-mail: [email protected]
Eur J Pediatr
bla OXA-58 and the intrinsic chromosomal bla OXA-51 carbapenemase genes. Conclusion: Active surveillance and enhanced infection control measures effectively contained spread of CRAB clone in the neonatal intensive care unit. Keywords Acinetobacter baumannii . Neonate . Outbreak . Carbapenem . Carbapenemase Abbreviations CRAB ID MIC NICU PFGE PICU
Carbapenem-resistant Acinetobacter baumannii Infectious disease Minimum inhibitory concentration Neonatal intensive care unit Pulsed-field gel electrophoresis Pediatric intensive care unit
Hippokration Hospital of Thessaloniki. This is a 920-bed tertiary-care general hospital treating both adult and paediatric patients. Apart from the NICU, there is a paediatric and an adult intensive care unit. The specific level 3 NICU has 44 beds including 15 intensive care beds providing critical care to neonates born in the same or other hospitals. In 2011, the NICU consisted of six wards, but there were no single rooms for isolation. A total of 674 neonates were admitted, while the occupancy rate was 80 % and the annual mortality rate 5 %. In this hospital, there are three infection control nurses and two infectious disease (ID) physicians as well as three ID fellows. Since 2011, there is active surveillance of four hospital-associated infections (bloodstream infections, urinary tract infections, lower respiratory tract infections and surgical site infections) caused by carbapenem-resistant bacteria [18]. Contact precautions are standard for patients with carbapenem-resistant bacteria. Study design
Introduction Acinetobacter baumannii has been increasingly recognized as an important nosocomial pathogen particularly in the intensive care settings [5]. The propensity of Acinetobacter spp. to grow at a range of different temperatures and pH values favours their dissemination in the hospital environment [5]. During the past 30 years, clones of Acinetobacter spp. have gradually accumulated resistance to penicillins, cephalosporins, quinolones and aminoglycosides while an increasing number of carbapenem-resistant Acinetobacter isolates have been reported worldwide [1, 32]. The emergence of the latter phenomenon is particularly concerning especially in neonates, which are very susceptible to infections and in which therapeutic options with antibiotics are quite limited. In the present study, an outbreak of carbapenem-resistant A. baumannii (CRAB) infections in a Greek neonatal intensive care unit (NICU) is reported. During the last months of 2011, an increased number of infections due to CRAB were noticed. The severe clinical manifestations caused by these isolates and their resistance phenotype prompted an urgent and intense investigation. Herein, we analyze the characteristics of the patients infected or colonized by CRAB and describe the methods used in order to investigate and successfully control the outbreak caused by a single clone of CRAB containing blaOXA-58 carbapenemase gene.
Patients and methods Clinical setting The outbreak occurred at the 1st Department of Neonatology and NICU of the Aristotle University located in the
This was a prospective study including the identification and investigation of the outbreak as well as the implementation of all bundle’s elements used for the management of this outbreak. The investigation was considered to be an infection control response, and as such, it was not subjected to approval by an institutional review board or required written informed consent from patients. Identification and investigation of the outbreak Prior to September 2011, no CRAB infections had occurred in the NICU. In September 2011, the first case of CRAB bloodstream infection occurred in a neonate on day 6 of hospitalization. After the occurrence of a second case of CRAB infection, an outbreak team was formed and consisted of two infection control nurses, the head of the microbiology department and paediatric ID physicians, all of whom worked together with the neonatologists and the other NICU staff. Case patients were defined as neonates with a nosocomial infection caused by a CRAB isolate identified in a clinically significant culture. Nosocomial infection was defined according to the criteria used by the Centers for Disease Control and Prevention [12]. A CRAB isolate was defined as having a minimum inhibitory concentration (MIC) of imipenem equal to or higher than 8 mg/L according to the interpretative criteria of the Clinical and Laboratory Standards Institute. This included intermediately resistant isolates [15]. Environmental cultures from places such as monitors, tables, patient charts and mechanical ventilation equipment were obtained, and a case-control study was conducted in order to identify possible risk factors associated with CRAB infection. For this purpose, cases were matched to controls (1:2 ratio), which were randomly selected among neonates
Eur J Pediatr
receiving intensive care during the same period for at least 48 h but had no positive cultures with CRAB. Demographic and clinical data recorded included date of birth, sex, gestational age, birth weight, underlying disease, invasive procedures, antimicrobial therapies and clinical outcome. Patient location and isolate assignment were recorded in the unit daily to determine potential modes of CRAB spread. Data collection and weekly epidemiologic surveillance were initiated shortly after the outbreak was noted.
Active surveillance for CRAB colonization
Implementation of all bundle’s elements
Temporary closure of the NICU
As soon as the outbreak was identified and investigation started, a bundle of actions were implemented as follow: Apart from the formation of the outbreak management team, the case-control study and the environmental cultures, other measures included the following: intensification of infection control measures, implementation of active surveillance cultures for CRAB colonization in hospitalized neonates, optimization of antimicrobial therapy and closure of the department to new admissions. Specifically, no neonates were admitted for 12 days; then, the department was admitting only neonates born in the same hospital for an additional week.
Temporary discontinuation of accepting new admissions to the NICU was decided after the results of the first active surveillance cultures during the 12th week of this outbreak.
Intensified infection control measures
Microbiologic methods and molecular typing
Intensification of infection control measures included the following: (1) Cohorting of neonates colonized or infected with CRAB in designated areas of the NICU. The newly admitted neonates were separated from the infected or colonized ones and managed by different medical and nursing teams (staff cohorting); (2) monitoring of adherence to contact precautions applied to neonates with culture results positive for CRAB and to standard precautions for all other neonates; (3) reevaluation and improvement of environmental cleaning and decontamination as well as the sterilization process of reusable equipment and identification of potential flaws in standard NICU practices. Processes such as intravenous access, use of heparin locks, preparation of intravenously administered fluids and use of syringes were reviewed and directly observed by the infection control team; (4) availability of replaceable dispensers containing alcohol hand sanitizer in all places for immediate access such as incubators, trolleys, nurse station and walls was regularly checked; (5) monitoring and evaluation of the adherence to infection control measures by the infection control nurses at various times of the day and (6) education of all NICU staff and neonates’ families by the ID physicians and infection control nurses on the importance of strict adherence to all enhanced infection control measures.
All clinical samples regularly obtained as clinically indicated for culture from all patients in the unit were processed using standard practice and culture media. Blood cultures were processed using the BacT/ALERT automated system (bioMérieux, Hazelwood, MO). Identification and susceptibility testing of A. baumannii isolates were performed by the microbiology laboratory using the Vitek 2 automated system (bioMérieux). The MIC determination and interpretation results from the automated system were comparable to those of the Clinical and Laboratory Standards Institute [15]. Imip en em MICs we re a lso c onf irmed by E- test (bioMérieux). The active surveillance specimens were inoculated in MacConkey agar plates supplemented with imipenem (1 mg/L). After overnight incubation at 37 °C, suspected Acinetobacter spp. colonies were further examined. Identification to the species level and antimicrobial susceptibility testing were performed as described above. Genes encoding metallo-β-lactamases (blaVIM, blaIMP and bla SIM ) and oxacillinase (OXA)-type carbapenemases (blaOXA51-like, blaOXA23-like, blaOXA24-like and blaOXA58-like) were sought by PCR amplification using primers and conditions previously published [10, 34]. For the genotypic analysis of the isolates, chromosomal DNA was digested using the ApaI restriction endonuclease (Fermendas, Athens, Greece), and the generated fragments were analyzed by pulsed-field gel
Active surveillance for CRAB colonization was started on the 12th week of the outbreak and continued until the 21st week. All neonates hospitalized in the NICU were eligible. More specific, perianal or stool samples were collected weekly from all neonates until discharge. Colonization with CRAB was defined as the acquisition of this pathogen during hospitalization (at least one positive surveillance culture) but without any evidence of clinical disease.
Changes in antibiotic policy First-line empirical antibiotic therapy including ampicillin with gentamicin was not modified during this outbreak. For the treatment of late-onset sepsis, both antibiotic regimen and duration were discussed daily with ID physicians. Colistin and carbapenems were administered only after ID approval during the outbreak period.
Eur J Pediatr
electrophoresis (PFGE) in a CHEF-DR III apparatus (Bio-Rad Laboratories, Athens, Greece) at 14 °C with a running time of 21 h and pulse times ranging from 3 to 40 s [27]. Banding patterns were compared visually according to the criteria suggested by Tenover et al. [29]. Clinical outcomes Daily prevalence of neonates infected by CRAB was calculated throughout the outbreak period. After the implementation of active surveillance cultures, daily prevalence of all infected or colonized neonates by CRAB was also assessed. All cause- and infection- related mortality was monitored during the outbreak period. Statistical analysis All continuous variables were expressed as mean±SEM (standard error of mean). In case-control study, Student’s t test or Mann-Whitney U test was used for the comparison of normally distributed and non-normally distributed continuous variables, respectively. For categorical variables, the Fisher’s exact test was used in univariate analysis. A two tailed p value of
