Infection DOI 10.1007/s15010-014-0654-9
CLINICAL AND EPIDEMIOLOGICAL STUDY
Recurrent outbreaks of Serratia marcescens among neonates and infants at a pediatric department: an outbreak analysis B. Iva´dy • D. Szabo´ • I. Damjanova ´ . Kenesei M. Pataki • M. Szabo´ • E
•
Received: 15 April 2014 / Accepted: 21 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Serratia marcescens is a known cause of bloodstream infections (BSIs) and outbreaks in neonates receiving intensive care. Our aim was to analyze clinical and epidemiological characteristics of two outbreaks detected in our unit to prevent and control further epidemic infections. Methods Two episodes of BSI outbreaks in neonates have been investigated in a 20-month period at a pediatric department of a medical university in Hungary. We collected all S. marcescens strains that were isolated in the study period, and two strains that were isolated before the B. Iva´dy (&) Department of Anaesthesiology and Intensive Care, Heim Pa´l ¨ ll} Children’s Hospital, U oi u´t 86, 1089 Budapest, Hungary e-mail: [email protected] B. Iva´dy D. Szabo´ Institute of Medical Microbiology, Semmelweis University, Budapest, Hungary e-mail: [email protected] I. Damjanova Department of Phage and Molecular Typing, National Center for Epidemiology, Budapest, Hungary e-mail: [email protected] M. Pataki M. Szabo´ Neonatal Intensive Care Unit, 1st Department of Pediatrics, Semmelweis University, Budapest, Hungary e-mail: [email protected] M. Szabo´ e-mail: [email protected] E´. Kenesei Microbiology Laboratory, 1st Department of Pediatrics, Semmelweis University, Budapest, Hungary e-mail: [email protected]
outbreaks. Strains were analyzed by pulsed-field gel electrophoresis (PFGE). Clinical data were collected for the BSIs during and between the outbreaks (n = 14). Results Out of the 28 S. marcescens isolates investigated by PFGE, 16 were blood isolates. All isolates represented four PFGE types. Pathogenic strains that caused epidemic BSIs were related to a single PFGE type (SM009). Strains with the same pulsotype could be detected before, between, and after the outbreak periods from surveillance cultures of neonates, and a water tap in the infant care unit despite intensive infection control measures. Case fatality rate of BSIs was 29 %. Rate of complications in central nervous system was high: 3/14 neonates developed meningitis. Conclusions Rapid spread and high mortality rate of S. marcescens infections necessitate a high suspicion when isolating this species in neonatal intensive care. Early identification of outbreaks is essential, that can be facilitated by determination of clonal relatedness using molecular methods, and with regular surveillance cultures of patients and environment. Keywords Serratia marcescens Outbreak Neonate Meningitis Pulsed-field gel electrophoresis
Introduction Serratia marcescens is a Gram-negative pathogen that has been recognized as a cause of hospital acquired infections and outbreaks of the compromised hosts in the last decades [1]. S. marcescens infections have been reported both in adult and pediatric patients undergoing invasive procedures; however, preterm infants and neonates seem to be one of the most threatened populations. Indeed, clusters of infections due to this bacterium have been associated with
123
B. Iva´dy et al.
high mortality and prominently high rate of central nervous system (CNS) involvement (e.g., meningitis and cerebral abscess) in neonatal and pediatric intensive care units (NICUs and PICUs, respectively) [2]. Reports of outbreaks of S. marcescens in neonatal and pediatric wards have been published in increasing number since the late 1990s. These reports revealed diverse sources of the infections [3–6]. Case–control studies identified several risk factors for acquisition of S. marcescens infection which included exposure to a single health care worker (HCW), length of NICU stay, maternal infection or chorioamnionitis, very low birth weight, and others [7–12]. A review article summarized observations about nosocomial infections in NICUs and PICUs in 2010 [2]. The main conclusions of these papers point out that the outbreaks may be successfully stopped with rapid detection and prompt intervention (i.e., infection control measures) even when the source of the infections cannot be identified. In this paper, we report a recurrent outbreak caused by S. marcescens in neonates and infants at different wards of a pediatric department. Aim of the outbreak analysis was to describe the spread of the epidemic S. marcescens strain and understand how it persisted and caused a recurrent outbreak in our clinic despite intensive outbreak control measures. These data can further increase our knowledge about this important clinical, microbiological and epidemiological issue.
Methods
marcescens strain during the investigated period were defined as an outbreak case. Epidemic strains were identified according to the PFGE typing described below. To understand the spread and persistence of S. marcescens at the wards, we collected all patients that were infected or colonized with S. marcescens in the study period by reviewing all S. marcescens strains isolated in our microbiology laboratory. Bacterial isolates and identification The clinical samples and surveillance samples were investigated at the Microbiology Laboratory of the 1st Department for Pediatrics and at Institute of Medical Microbiology, Semmelweis University, Budapest. Blood culture tests were performed using BacT/Alert system (bioMerieux). For the identification of S. marcescens strains API20E test system (bioMerieux) was used. Antibiotic susceptibility test The minimal inhibitory concentrations (MICs) were determined by microdilution method for the following antibiotics: amoxicillin/clavulanic acid, piperacillin/tazobactam, ceftazidime, ceftriaxone, cefepime, ertapenem, imipenem, meropenem, gentamicin, tobramycin, amikacin and ciprofloxacin. Escherichia coli ATCC 25922 was used as the reference strain for antimicrobial susceptibility testing. All susceptibility results were interpreted using the breakpoints of the European Committee on Antimicrobial Susceptibility Testing (EUCAST).
Setting description Pulsed-field gel electrophoresis (PFGE) The outbreaks have been detected at the first Department of Pediatrics of Semmelweis University, Budapest, Hungary. Patients of NICU, PICU, Neonatal Surgery Unit (NSU), and Infant Special Care Unit (ISCU) have been involved in the clusters of nosocomial infections. The number of beds of the units is 25, 8, 9, and 17, respectively. The number of patients treated in these units is approximately 2,000–2,500 infants per year. The affected units are on the same floor of the building, except ISCU, which is located in another building. Patients are transferred from one to another unit according to the changes in conditions or different need for care. Each nurse works on the same ward and is responsible for 1–3 patients per shift. However, same physicians are responsible for patients at NICU, PICU, NSU and ISCU as well.
The clonal relationships of the S. marcescens isolates were studied by pulsed-field gel electrophoresis (PFGE) at the Department of Phage and Molecular typing, National Center for Epidemiology. The PFGE method was performed in line with the standardized CDC protocol [13]. Gels were interpreted with Fingerprinting II Informatix Software (BioRad). Levels of similarity were calculated with the Dice coefficient, and UPGMA (‘unweighted pair group method with arithmetic averages’) was used for the cluster analysis of the PFGE patterns. Pulsotypes (PTs) were defined at 85 % similarity between macrorestriction patterns and marked by letters and numbers according to the criteria established by Tenover et al. [14]. Clonally related isolates were supposed if they belonged to the same PT.
Outbreak analysis: case definition and patient selection
Ethical issues
In the outbreak analysis any neonates and infants with bloodstream infection (BSI) caused by an epidemic of S.
A prospective study from 2009 to 2012 involving all bloodstream infections caused by Gram-negative bacteria
123
Recurrent outbreaks of S. marcescens among neonates and infants Table 1 Characteristics of S. marcescens isolates Patient
Date of detection
Source
PFGE type
Control 1
Day-136
Blood
SM014
Control 2
Day-127
Blood
SM009
Case 1a
Day 0
Blood
SM009
Carrier 1
Day 1
Cenral venous catheter
SM009
Case 2a
Day 4
Blood
SM009
Case 3
Day 4
Blood
SM009
Case 4
Day 4
Blood
SM009
Carrier 2
Day 19
Throat
SM009
Carrier 3
Day 24
Anus
SM010
Carrier 4
Day 31
Trachea
SM009
Case 5a
Day 33
Blood
SM009
Case 6
Day 33
Blood
SM009
Case 5b
Day 36
Cerebrospinal fluid
SM009
Case 7
Day 36
Blood
SM009
Case 8
Day 38
Blood
SM009
Case 9
Day 38
Blood
SM009
Carrier 5 Sporadic 1
Day 41 Day 93
Nose Blood
SM009 SMV1a
Carrier 5
Day 164
Wound
SM009
Environment
Day 201
Water tap
SM009
Outbreak period 1
Results Outbreak description: cases and clinical isolates
Carrier 6
Day 225
Anus
SM009
Sporadic 2
Day 266
Blood
SM015
Day 487
Blood
SM009
Carrier 7
Day 501
Nose
SM009
Case 11
Day 504
Blood
SM009
Case 12
Day 507
Blood
SM009
Carrier 8
Day 557
Anus
SM009
Carrier 9
Day 561
Anus
SMV1
Carrier 10
Day 563
Anus
SMV1
Carrier 11
Day 585
Anus
SM009
Outbreak period 2 Case 10c
all infants whose clinical data have been used in this study.
‘Day 0’ indicates the day when the first epidemic BSI was detected Strains in bold type are blood isolates of SM009 (outbreak) pulsotype Strains in italic type are sporadic blood isolates a
Neonates with BSI and central nervous system infection
b
This strain has been isolated from the cerebrospinal fluid of Case 5
c
This neonate has been admitted from another neonatal unit of Budapest, and blood culture was positive in 24 h from admission
in five pediatric centers in Budapest, Hungary, has been approved by the Scientific and Research Ethics Council of the Hungarian Medical Research Council under the reference number 845-0/2010-1018EKU (39/PI/010). Clinical data collection during this outbreak analysis has been conducted as part of this multicenter study. Informed consent has been attained from the parents of
From May to June 2010 and August to September 2011 we investigated two outbreaks in a pediatric department. In all, we detected 12 neonates with nosocomial S. marcescens BSIs, and two blood isolates were detected between the two outbreak periods as well. In addition, 13 isolates causing colonization were investigated. Two S. marcescens strains isolated before the outbreak periods were also collected, aiming to compare these strains with the strains isolated during the outbreaks. Table 1 lists the investigated S. marcescens strains. During the outbreak period 1 nine clinical infections and six patients colonized with S. marcescens have been detected in the NICU from the first days of May until the middle of June 2010. The BSIs have been identified in two consecutive clusters. In the first cluster four premature neonates developed BSI at the NICU on the ‘Day 0’ and ‘Day 4’ of the outbreak, as shown in Table 1 (Cases 1–4). In the second cluster that appeared in early June (between Day 33 and Day 38 of the outbreak), five neonates had positive blood cultures (Cases 5–9). In case of one neonate (Case 5), the same strain could be detected from the cerebrospinal fluid, as well. During this second cluster infections were found not only at the NICU, but in the PICU, NSU and ISCU as well. During outbreak period 1 colonization could not be demonstrated in neonates that developed BSI. However, five other asymptomatic infants have been identified as a carrier of the pathogen (Carriers 1–5). These isolates originated from the nose, throat, trachea, anus and central venous catheter tip sample, as shown in Fig. 1. Between June 2010 to August 2011 two infants developed S. marcescens BSI at NSU sporadically (indicated as Sporadic 1 and Sporadic 2 in Table 1), and two cases of colonization have been found. In November 2010 one strain was isolated from an environmental sample of the ISCU, as described later. The outbreak strain was isolated from one infant (Carrier 5) two separate times: on Day 41 and on Day 164 as well. Outbreak period 2 started at the end of August 2011 (Day 487 in Table 1). Three cases with BSIs and two infants with colonization have been found in the NICU and in the PICU (Cases 10–12 in Table 1). The first neonate with BSI during this period (Case 10) was admitted from a neonatal ward of another hospital in Budapest, and blood culture became positive within 24 h from admission. One infant had eyecolonization besides the positive blood culture.
123
B. Iva´dy et al.
Fig. 1 Chronology of Serratia marcescens infections and colonization during the investigated period. This figure demonstrates that the epidemic PFGE genotype (SM009) could be detected during the whole study period. During the outbreaks SM009 caused all
bloodstream infections (black boxes). Between the outbreaks two patients was found to be colonized with the SM009 strain, and the SM009 type could be isolated from one water tap sample culture as well
After the second outbreak four children have been detected to be colonized with S. marcescens in November and December 2011. However, until the end of the investigated period, no more invasive infections have been observed.
solutions, sinks, ventilator and gas delivery devices, incubators) have been periodically cultured during and between the outbreaks. Hand cultures of HCWs were also obtained. During the outbreak periods altogether 326 environmental specimens were cultured. The only positive environmental culture was an epidemic S. marcescens strain isolated from a water tap in the ISCU between the outbreak periods, in November 2010 (indicated as ‘Environment’ on ‘Day 201’ in Table 1).
Outbreak control measures In May 2010 infection control staff including neonatologists, infectious diseases specialist, microbiologist, epidemiologist and infection control nurse started interventions to stop the outbreak. Education of hand hygiene guidelines has been started for all HCWs of the department; more attention was kept on cohort care and disinfection of environment. Implementation and adherence to the guidelines were regularly re-evaluated. As part of the intervention infected and colonized patients were transferred to an isolated ward, with dedicated nursing team and cleaning staff. Environmental culture survey Besides routine surveillance cultures from patients, various environmental samples (breast milk, formula, injection drugs, infusions, components of parenteral nutrition, disinfection
123
Clinical infections: presentation and outcome Clinical data about 12 S. marcescens BSIs from the investigated outbreak periods and two BSIs detected between the outbreak periods were collected. Data are summarized in Table 2. The babies’ age at the time of the positive blood culture was in the range of 5–165 days (median 37 days). The time from admission to the pediatric department until the development of infection was 1–110 days (median 24 days). The observed neonates were born at a median gestational age of 33.5 weeks (range 24–37 weeks) and the median birth weight was 2,070 g (range 600–3,450 g). Five (36 %) of all babies had an extremely low birth weight (ELBW). Three of the 14 infants (21 %) developed neuroinfection during the course of S. marcescens sepsis; the microbe could be isolated
NICU
NICU
NICU
NICU
Infant care unit
PICU
Infant care unit/PICU Infant care unit
Case 2
Case 3
Case 4
Case 5
Case 6
Case 7
Case 8
Neonatal surgery/PICU
Sporadic 2
NICU
NICU
Case 11
Case 12
83
43
5
59
110
11
92
51
165
25
5
21
31
10
Age (days)
25
25
37
34
33
35
36
37
34
33
24
24
31
34
Gestational age (weeks)
670
630
2,280
1,990
2,180
2,800
2,150
3,450
3,220
2,180
990
600
990
1,900
Birth weight (g)
Prematurity, BPD, ROP
Prematurity, IRDS
IUGR
Prematurity, tracheoesophageal fistula, pneumonia
Prematurity, jejunal atresia
Prematurity, dehydration
Smith–Lemli–Opitz syndrome
Undetermined CNS disease
Intestinal lymphangiectasia, otitis media
Prematurity, IRDS
Prematurity, IUGR, IRDS, NEC, PDA, IVH
Prematurity, IUGR, IRDS, NEC, PDA
Prematurity, IUGR, IRDS, NEC, PDA
Prematurity, IRDS
Underlying condition
18
43
1
54
110
3
32
4
2
24
24
24
31
10
In hospital days to BSI
Blood
Blood
Blood ? eye
Blood ? trachea
Blood
Blood
Blood
Blood
Blood
Blood and CSF
Blood
Blood
Blood
Blood and CSF
Cultures of S. marcescens
Sepsis
Sepsis
Sepsis
Sepsis
Sepsis
Bacteremia
Sepsis
Sepsis
Sepsis
Sepsis, meningitis
Sepsis
Sepsis
Sepsis, meningitis
Sepsis, meningitis, brain abscess
Clinical manifestation
Alive
Alive
Alive
Died
Died
Alive
Alive
Alive
Alive
Died
Died
Alive
Alive
Alive
28-day outcome
SM009
SM009
SM009
SM015
V1a
SM009
SM009
SM009
SM009
SM009
SM009
SM009
SM009
SM009
PFGE type
NICU Neonatal Intensive Care Unit, PICU Pediatric Intensive Care Unit, IRDS infantile respiratory distress syndrome, IUGR intrauterine growth retardation, NEC necrotizing enterocolitis, PDA patent ductus arteriosus, CNS central nervous system, BPD bronchopulmonary dysplasia, ROP retinopathy of prematurity
PICU
Case 10
Outbreak period 2
Neonatal surgery/PICU
Sporadic 1
Case 9
NICU
Hospital unit
Case 1
Outbreak period 1
Patient
Table 2 Epidemiological and clinical data of neonates with S. marcescens bloodstream infection (BSI)
Recurrent outbreaks of S. marcescens among neonates and infants
123
B. Iva´dy et al.
from the cerebrospinal fluid in two patients (both patients had positive blood cultures as well). Among the neonates with neuroinfection only one was an ELBW premature. Among low birth weight and premature neonates the underlying conditions were mostly IRDS, IUGR, PDA. Three out of four babies that died had more severe conditions and/or congenital malformations (intraventricular hemorrhage, jejunal atresia, and tracheoesophageal fistula). The 28-day mortality was found to be 29 % (4 of 14 cases), four of the five ELBW infants recovered from the S. marcescens BSI. Antimicrobial resistance All invasive isolates had the same antimicrobial resistance pattern. The strains causing BSIs were resistant to amoxicillin, amoxicillin/clavulanic acid, cefuroxime, and showed in vitro sensitivity to ceftriaxone, ceftazidime, cefepime, ciprofloxacin, trimethoprim/sulfamethoxazol, aminoglycosides (tobramycin, gentamicin, amikacin), pipercillin/tazobactam, imipenem and meropenem. Among strains that caused colonization, one isolate from the anus of a neonate was found to be resistant against ceftriaxone, piperacillin/tazobactam, carbapanems, aminoglycosides, and was sensitive only to ceftazidime, ciprofloxacin and levofloxacin. This strain had different PFGE type than any other strains isolated in the investigated period (see below). Molecular (PFGE) analysis PFGE analysis was performed for 16 BSI isolates, for one CSF isolate (from one patient that had a blood culture isolate as well), for 12 colonizing strains, one strain isolated from a water tap, and for two strains obtained from sporadic BSIs detected 6 months before the outbreak (these strains were retrospectively analyzed). Table 1 lists the S. marcescens isolates that have undergone PFGE analysis. In all, four different PFGE patterns have been found in our study: the SM009, SM010, SM015, SMV1/V1a. In our observation all blood culture strains isolated during the two outbreaks belonged to the SM009 PFGE pattern. The blood isolates detected between the outbreak periods represented two different PFGE types. Colonizing strains in this period belonged to the SM009 pulsotype and the strain isolated from a water tap between the outbreak periods proved to be SM009 as well, molecular investigations revealed that the strain responsible for the outbreaks (SM009) could be detected before, between and after the outbreaks at the pediatric department (Fig. 1). Discussion Serratia marcescens, a Gram-negative bacterium, is a wellknown pathogen of nosocomial outbreaks, especially in ICUs. Epidemiological, microbiological and clinical
123
characteristics of outbreaks in NICUs and PICUs have been intensively studied in the last decade. We report two consecutive outbreaks of S. marcescens BSIs in neonates and infants in three different wards of a pediatric center. We identified four different PFGE genotypes from which one was responsible for all BSIs during the two outbreak periods, but not for the sporadic BSIs between the outbreaks. The source of the infections could be identified with environmental cultures in several outbreaks reported recently. S. marcescens strains could be isolated from the hands of HCWs [5, 9, 15–17], hand washing soaps [15, 18, 19], sink drain [20], incubator [5], laryngoscope [4, 21], different drugs [22], parenteral nutrition [23], and baby shampoo [24]. During the outbreaks we could not detect the epidemic S. marcescens strain (PFGE type SM009) in the environment of the patients. However, this strain was isolated as a colonizing agent in asymptomatic patients during the outbreak periods, and detected in one culture of a water tap sample between the two outbreaks. We suppose that besides an environmental focus, colonized patients may serve as reservoirs and source for recurrent infections among NICU patients, as described earlier by many authors [25–28]. In our view, regular screening for specific pathogens to detect colonization is of particular importance in NICUs. For example, weekly screening for Gram-negative pathogens can help empiric antibiotic therapy [29] and identify specific pathogens (including Serratia) that are able to cause epidemics and recurrent infections in these settings. In our clinic’s NICU weekly screening cultures of anal specimens for multidrug resistant bacteria and ESBLproducing Gram-negatives have been performed regularly in all patients since 2008. In case of an outbreak or cluster of infections screening policy is extended to the causative organism (e.g., vancomycin-resistant enterococci or methicillin-resistant Staphylococcus aureus). During the periods of the described S. marcescens outbreaks in 2010–2011, surveillance has been extended to this pathogen, and altogether 3,386 screening cultures has been performed in these time periods. Results showed that during the whole investigated period there were almost always neonates in the NICU colonized with SM009 strain. However, only one environmental sample proved the presence of the pathogen, which was isolated in a period when no invasive infections were caused by this strain. Importantly, colonization with SM009 disappeared in January 2011, supposedly due to the infection control measures that stopped the transmission of the pathogenic strain. It is important that the first neonate with BSI in the outbreak period 2 was admitted from another hospital, and the blood culture was positive within 24 h from admission, suggesting that the SM009 was probably re-introduced to the pediatric clinic from the referring hospital.
Recurrent outbreaks of S. marcescens among neonates and infants
It would be useful to conduct epidemiological studies in other hospital settings in the same region to detect the predominant strains or clones that spread from hospital to hospital and ward to ward. Isolation of colonized and/or infected patients and barrier precautions are needed to control the horizontal spread of pathogens among patients in the same setting. During the outbreak periods in our institution, neonates colonized or infected by S. marcescens were isolated in a single box, managed by dedicated nurses, and using gloves, gowns and masks was obligatory for those who were caring for these patients. When a colonized patient was transferred to another ward, isolation and barrier precautions were strictly continued. Importantly, asymptomatic colonized patients were not treated with antibiotics. Recognizing that physicians may be an epidemiological link between the different wards, physicians were educated about nosocomial transmission of pathogens, and regular hand-hygiene educations were performed for all HCWs. In our observation, during outbreak period 1, the rate of neuroinfections (i.e., meningitis or brain abscess) as complication or focus of septicemia was high (33.3 %). Two neonates had meningitis, and one developed brain abscess, which is a known, but rare entity, as described by Hirooka et al. and Messerschmidt et al. [30, 31]. The rate of neuroinfections during neonatal S. marcescens outbreaks is 0–60 %. In a systematic analysis that reviewed 27 outbreaks involving 239 cases the rate was reported to be 7 % [2]. The highest rate was observed by Berger et al. [32], who described five cases of S. marcescens infections, from which three developed meningoencephalitis with brain abscess. Madani et al. [24] reported that the rate of neuroinfections during an outbreak in an NICU was 14 % (2 meningitis of 14 infected patients). Bizzarro et al. [33] analyzed sporadic neonatal S. marcescens infections compared to E. coli infections, and the rate of neuroinfections was found to be 24 and 7 %, respectively. The high rate of CNS involvement necessitates early treatment with antibiotics that are able to cross the blood–brain barrier in all neonates with S. marcescens infections, and early diagnostic measures (lumbar puncture, neuroimaging studies) to detect meningitis and/or brain abscess.
Conclusions Serratia marcescens can cause severe nosocomial outbreaks in neonates and infants. PFGE is a useful method to identify the predominant strain that may cause recurrent outbreaks. The outbreak strain can be detected in asymptomatic patients and in the environment without causing invasive disease in that period of time, but it can be the source of outbreaks later. Regular surveillance
cultures are needed even when the outbreak is controlled by isolation of infected patients, strict hand hygiene policies and continuous education. Early antibiotic therapy is needed in all cases of neonatal S. marcescens infections, always considering the possibility of CNS involvement. Acknowledgments We are very grateful to Natasa Pesti, Kla´ra To´th and Ma´ria Sze´na´si for their excellent assistance. The work was supported by Hungarian National Scientific Research Fund: OTKA108481. Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
References 1. Hejazi A, Falkiner FR. Serratia marcescens. J Med Microbiol. 1997;46(11):903–12. 2. Voelz A, Muller A, Gillen J, Le C, Dresbach T, Engelhart S, et al. Outbreaks of Serratia marcescens in neonatal and pediatric intensive care units: clinical aspects, risk factors and management. Int J Hyg Environ Health. 2010;213(2):79–87. doi:10.1016/ j.ijheh.2009.09.003. 3. Cimolai N, Trombley C, Wensley D, LeBlanc J. Heterogeneous Serratia marcescens genotypes from a nosocomial pediatric outbreak. Chest. 1997;111(1):194–7. 4. Jones BL, Gorman LJ, Simpson J, Curran ET, McNamee S, Lucas C, et al. An outbreak of Serratia marcescens in two neonatal intensive care units. J Hosp Infect. 2000;46(4):314–9. doi:10. 1053/jhin.2000.0837. 5. Jang TN, Fung CP, Yang TL, Shen SH, Huang CS, Lee SH. Use of pulsed-field gel electrophoresis to investigate an outbreak of Serratia marcescens infection in a neonatal intensive care unit. J Hosp Infect. 2001;48(1):13–9. doi:10.1053/jhin.2001.0947. 6. Villa J, Alba C, Barrado L, Sanz F, Del Castillo EG, Viedma E, et al. Long-term evolution of multiple outbreaks of Serratia marcescens bacteremia in a neonatal intensive care unit. Pediatr Infect Dis J. 2012;31(12):1298–300. doi:10.1097/INF. 0b013e318267f441. 7. Lai KK, Baker SP, Fontecchio SA. Rapid eradication of a cluster of Serratia marcescens in a neonatal intensive care unit: use of epidemiologic chromosome profiling by pulsed-field gel electrophoresis. Infect Control Hosp Epidemiol. 2004;25(9):730–4. doi:10.1086/502468. 8. Maragakis LL, Winkler A, Tucker MG, Cosgrove SE, Ross T, Lawson E, et al. Outbreak of multidrug-resistant Serratia marcescens infection in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2008;29(5):418–23. doi:10.1086/587969. 9. Friedman ND, Kotsanas D, Brett J, Billah B, Korman TM. Investigation of an outbreak of Serratia marcescens in a neonatal unit via a case-control study and molecular typing. Am J Infect Control. 2008;36(1):22–8. doi:10.1016/j.ajic.2006.12.012. 10. Al Jarousha AM. El Qouqa IA, El Jadba AH, Al Afifi AS. An outbreak of Serratia marcescens septicaemia in neonatal intensive care unit in Gaza City, Palestine. J Hosp Infect. 2008;70(2):119–26. doi:10.1016/j.jhin.2008.06.028. 11. Buffet-Bataillon S, Rabier V, Betremieux P, Beuchee A, Bauer M, Pladys P, et al. Outbreak of Serratia marcescens in a neonatal intensive care unit: contaminated unmedicated liquid soap and risk factors. J Hosp Infect. 2009;72(1):17–22. doi:10.1016/j.jhin. 2009.01.010.
123
B. Iva´dy et al. 12. Iosifidis E, Farmaki E, Nedelkopoulou N, Tsivitanidou M, Kaperoni M, Pentsoglou V, et al. Outbreak of bloodstream infections because of Serratia marcescens in a pediatric department. Am J Infect Control. 2011;. doi:10.1016/j.ajic.2011.03.020. 13. CDC. Standardized molecular subtyping of foodborne bacterial pathogens by pulsed-field gel electrophoresis: training manual. (Centers for disease control and prevention, Atlanta, GA). 2000. 14. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol. 1995;33(9):2233–9. 15. Manning ML, Archibald LK, Bell LM, Banerjee SN, Jarvis WR. Serratia marcescens transmission in a pediatric intensive care unit: a multifactorial occurrence. Am J Infect Control. 2001;29(2):115–9. doi:10.1067/mic.2001.114222. 16. Milisavljevic V, Wu F, Larson E, Rubenstein D, Ross B, Drusin LM, et al. Molecular epidemiology of Serratia marcescens outbreaks in two neonatal intensive care units. Infect Control Hosp Epidemiol. 2004;25(9):719–21. doi:10.1086/502466. 17. Villari P, Crispino M, Salvadori A, Scarcella A. Molecular epidemiology of an outbreak of Serratia marcescens in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2001;22(10):630–4. doi:10.1086/501834. 18. Rabier V, Bataillon S, Jolivet-Gougeon A, Chapplain JM, Beuchee A, Betremieux P. Hand washing soap as a source of neonatal Serratia marcescens outbreak. Acta Paediatr. 2008;97(10):1381–5. doi:10.1111/j.1651-2227.2008.00953.x. 19. Polilli E, Parruti G, Fazii P, D’Antonio D, Palmieri D, D’Incecco C et al. Rapidly controlled outbreak of Serratia marcescens infection/colonisations in a neonatal intensive care unit, Pescara General Hospital, Pescara, Italy, April 2011. Euro Surveill. 2011;16(24):16–18. 20. Maltezou HC, Tryfinopoulou K, Katerelos P, Ftika L, Pappa O, Tseroni M, et al. Consecutive Serratia marcescens multiclone outbreaks in a neonatal intensive care unit. Am J Infect Control. 2012;40(7):637–42. doi:10.1016/j.ajic.2011.08.019. 21. Cullen MM, Trail A, Robinson M, Keaney M, Chadwick PR. Serratia marcescens outbreak in a neonatal intensive care unit prompting review of decontamination of laryngoscopes. J Hosp Infect. 2005;59(1):68–70. doi:10.1016/j.jhin.2004.08.003. 22. Fleisch F, Zimmermann-Baer U, Zbinden R, Bischoff G, Arlettaz R, Waldvogel K, et al. Three consecutive outbreaks of Serratia marcescens in a neonatal intensive care unit. Clin Infect Dis. 2002;34(6):767–73. doi:10.1086/339046. 23. Arslan U, Erayman I, Kirdar S, Yuksekkaya S, Cimen O, Tuncer I, et al. Serratia marcescens sepsis outbreak in a neonatal
123
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
intensive care unit. Pediatr Int. 2010;52(2):208–12. doi:10.1111/j. 1442-200X.2009.02934.x. Madani TA, Alsaedi S, James L, Eldeek BS, Jiman-Fatani AA, Alawi MM, et al. Serratia marcescens-contaminated baby shampoo causing an outbreak among newborns at King Abdulaziz University Hospital, Jeddah Saudi Arabia. J Hosp Infect. 2011;78(1):16–9. doi:10.1016/j.jhin.2010.12.017. Sarvikivi E, Lyytikainen O, Salmenlinna S, Vuopio-Varkila J, Luukkainen P, Tarkka E, et al. Clustering of Serratia marcescens infections in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2004;25(9):723–9. doi:10.1086/502467. Miranda-Novales G, Leanos-Miranda B, Diaz-Ramos R, Gonzalez-Tejeda L, Peregrino-Bejarano L, Villegas-Silva R, et al. An outbreak due to Serratia marcescens in a neonatal intensive care unit typed by 2-day pulsed field gel electrophoresis protocol. Arch Med Res. 2003;34(3):237–41. David MD, Weller TM, Lambert P, Fraise AP. An outbreak of Serratia marcescens on the neonatal unit: a tale of two clones. J Hosp Infect. 2006;63(1):27–33. doi:10.1016/j.jhin.2005.11.006. Steppberger K, Walter S, Claros MC, Spencker FB, Kiess W, Rodloff AC, et al. Nosocomial neonatal outbreak of Serratia marcescens––analysis of pathogens by pulsed field gel electrophoresis and polymerase chain reaction. Infection. 2002;30(5):277–81. doi:10.1007/s15010-002-2141-y. Simon A, Tenenbaum T. Surveillance of multidrug-resistant gram-negative pathogens in high-risk neonates—does it make a difference? Pediatr Infect Dis J. 2013;32(4):407–9. Hirooka TM, Fontes RB, Diniz EM, Pinto FC, Matushita H. Cerebral abscess caused by Serratia marcescens in a premature neonate. Arq Neuropsiquiatr. 2007;65(4A):1018–21 S0004282X2007000600021. Messerschmidt A, Prayer D, Olischar M, Pollak A, Birnbacher R. Brain abscesses after Serratia marcescens infection on a neonatal intensive care unit: differences on serial imaging. Neuroradiology. 2004;46(2):148–52. doi:10.1007/s00234-003-1140-8. Berger A, Rohrmeister K, Haiden N, Assadian O, Kretzer V, Kohlhauser C. Serratia marcescens in the neonatal intensive care unit: re-emphasis of the potentially devastating sequelae. Wien Klin Wochenschr. 2002;114(23–24):1017–22. Bizzarro MJ, Dembry LM, Baltimore RS, Gallagher PG. Casecontrol analysis of endemic Serratia marcescens bacteremia in a neonatal intensive care unit. Arch Dis Child Fetal Neonatal Ed. 2007;92(2):120–6.
CLINICAL AND EPIDEMIOLOGICAL STUDY
Recurrent outbreaks of Serratia marcescens among neonates and infants at a pediatric department: an outbreak analysis B. Iva´dy • D. Szabo´ • I. Damjanova ´ . Kenesei M. Pataki • M. Szabo´ • E
•
Received: 15 April 2014 / Accepted: 21 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Serratia marcescens is a known cause of bloodstream infections (BSIs) and outbreaks in neonates receiving intensive care. Our aim was to analyze clinical and epidemiological characteristics of two outbreaks detected in our unit to prevent and control further epidemic infections. Methods Two episodes of BSI outbreaks in neonates have been investigated in a 20-month period at a pediatric department of a medical university in Hungary. We collected all S. marcescens strains that were isolated in the study period, and two strains that were isolated before the B. Iva´dy (&) Department of Anaesthesiology and Intensive Care, Heim Pa´l ¨ ll} Children’s Hospital, U oi u´t 86, 1089 Budapest, Hungary e-mail: [email protected] B. Iva´dy D. Szabo´ Institute of Medical Microbiology, Semmelweis University, Budapest, Hungary e-mail: [email protected] I. Damjanova Department of Phage and Molecular Typing, National Center for Epidemiology, Budapest, Hungary e-mail: [email protected] M. Pataki M. Szabo´ Neonatal Intensive Care Unit, 1st Department of Pediatrics, Semmelweis University, Budapest, Hungary e-mail: [email protected] M. Szabo´ e-mail: [email protected] E´. Kenesei Microbiology Laboratory, 1st Department of Pediatrics, Semmelweis University, Budapest, Hungary e-mail: [email protected]
outbreaks. Strains were analyzed by pulsed-field gel electrophoresis (PFGE). Clinical data were collected for the BSIs during and between the outbreaks (n = 14). Results Out of the 28 S. marcescens isolates investigated by PFGE, 16 were blood isolates. All isolates represented four PFGE types. Pathogenic strains that caused epidemic BSIs were related to a single PFGE type (SM009). Strains with the same pulsotype could be detected before, between, and after the outbreak periods from surveillance cultures of neonates, and a water tap in the infant care unit despite intensive infection control measures. Case fatality rate of BSIs was 29 %. Rate of complications in central nervous system was high: 3/14 neonates developed meningitis. Conclusions Rapid spread and high mortality rate of S. marcescens infections necessitate a high suspicion when isolating this species in neonatal intensive care. Early identification of outbreaks is essential, that can be facilitated by determination of clonal relatedness using molecular methods, and with regular surveillance cultures of patients and environment. Keywords Serratia marcescens Outbreak Neonate Meningitis Pulsed-field gel electrophoresis
Introduction Serratia marcescens is a Gram-negative pathogen that has been recognized as a cause of hospital acquired infections and outbreaks of the compromised hosts in the last decades [1]. S. marcescens infections have been reported both in adult and pediatric patients undergoing invasive procedures; however, preterm infants and neonates seem to be one of the most threatened populations. Indeed, clusters of infections due to this bacterium have been associated with
123
B. Iva´dy et al.
high mortality and prominently high rate of central nervous system (CNS) involvement (e.g., meningitis and cerebral abscess) in neonatal and pediatric intensive care units (NICUs and PICUs, respectively) [2]. Reports of outbreaks of S. marcescens in neonatal and pediatric wards have been published in increasing number since the late 1990s. These reports revealed diverse sources of the infections [3–6]. Case–control studies identified several risk factors for acquisition of S. marcescens infection which included exposure to a single health care worker (HCW), length of NICU stay, maternal infection or chorioamnionitis, very low birth weight, and others [7–12]. A review article summarized observations about nosocomial infections in NICUs and PICUs in 2010 [2]. The main conclusions of these papers point out that the outbreaks may be successfully stopped with rapid detection and prompt intervention (i.e., infection control measures) even when the source of the infections cannot be identified. In this paper, we report a recurrent outbreak caused by S. marcescens in neonates and infants at different wards of a pediatric department. Aim of the outbreak analysis was to describe the spread of the epidemic S. marcescens strain and understand how it persisted and caused a recurrent outbreak in our clinic despite intensive outbreak control measures. These data can further increase our knowledge about this important clinical, microbiological and epidemiological issue.
Methods
marcescens strain during the investigated period were defined as an outbreak case. Epidemic strains were identified according to the PFGE typing described below. To understand the spread and persistence of S. marcescens at the wards, we collected all patients that were infected or colonized with S. marcescens in the study period by reviewing all S. marcescens strains isolated in our microbiology laboratory. Bacterial isolates and identification The clinical samples and surveillance samples were investigated at the Microbiology Laboratory of the 1st Department for Pediatrics and at Institute of Medical Microbiology, Semmelweis University, Budapest. Blood culture tests were performed using BacT/Alert system (bioMerieux). For the identification of S. marcescens strains API20E test system (bioMerieux) was used. Antibiotic susceptibility test The minimal inhibitory concentrations (MICs) were determined by microdilution method for the following antibiotics: amoxicillin/clavulanic acid, piperacillin/tazobactam, ceftazidime, ceftriaxone, cefepime, ertapenem, imipenem, meropenem, gentamicin, tobramycin, amikacin and ciprofloxacin. Escherichia coli ATCC 25922 was used as the reference strain for antimicrobial susceptibility testing. All susceptibility results were interpreted using the breakpoints of the European Committee on Antimicrobial Susceptibility Testing (EUCAST).
Setting description Pulsed-field gel electrophoresis (PFGE) The outbreaks have been detected at the first Department of Pediatrics of Semmelweis University, Budapest, Hungary. Patients of NICU, PICU, Neonatal Surgery Unit (NSU), and Infant Special Care Unit (ISCU) have been involved in the clusters of nosocomial infections. The number of beds of the units is 25, 8, 9, and 17, respectively. The number of patients treated in these units is approximately 2,000–2,500 infants per year. The affected units are on the same floor of the building, except ISCU, which is located in another building. Patients are transferred from one to another unit according to the changes in conditions or different need for care. Each nurse works on the same ward and is responsible for 1–3 patients per shift. However, same physicians are responsible for patients at NICU, PICU, NSU and ISCU as well.
The clonal relationships of the S. marcescens isolates were studied by pulsed-field gel electrophoresis (PFGE) at the Department of Phage and Molecular typing, National Center for Epidemiology. The PFGE method was performed in line with the standardized CDC protocol [13]. Gels were interpreted with Fingerprinting II Informatix Software (BioRad). Levels of similarity were calculated with the Dice coefficient, and UPGMA (‘unweighted pair group method with arithmetic averages’) was used for the cluster analysis of the PFGE patterns. Pulsotypes (PTs) were defined at 85 % similarity between macrorestriction patterns and marked by letters and numbers according to the criteria established by Tenover et al. [14]. Clonally related isolates were supposed if they belonged to the same PT.
Outbreak analysis: case definition and patient selection
Ethical issues
In the outbreak analysis any neonates and infants with bloodstream infection (BSI) caused by an epidemic of S.
A prospective study from 2009 to 2012 involving all bloodstream infections caused by Gram-negative bacteria
123
Recurrent outbreaks of S. marcescens among neonates and infants Table 1 Characteristics of S. marcescens isolates Patient
Date of detection
Source
PFGE type
Control 1
Day-136
Blood
SM014
Control 2
Day-127
Blood
SM009
Case 1a
Day 0
Blood
SM009
Carrier 1
Day 1
Cenral venous catheter
SM009
Case 2a
Day 4
Blood
SM009
Case 3
Day 4
Blood
SM009
Case 4
Day 4
Blood
SM009
Carrier 2
Day 19
Throat
SM009
Carrier 3
Day 24
Anus
SM010
Carrier 4
Day 31
Trachea
SM009
Case 5a
Day 33
Blood
SM009
Case 6
Day 33
Blood
SM009
Case 5b
Day 36
Cerebrospinal fluid
SM009
Case 7
Day 36
Blood
SM009
Case 8
Day 38
Blood
SM009
Case 9
Day 38
Blood
SM009
Carrier 5 Sporadic 1
Day 41 Day 93
Nose Blood
SM009 SMV1a
Carrier 5
Day 164
Wound
SM009
Environment
Day 201
Water tap
SM009
Outbreak period 1
Results Outbreak description: cases and clinical isolates
Carrier 6
Day 225
Anus
SM009
Sporadic 2
Day 266
Blood
SM015
Day 487
Blood
SM009
Carrier 7
Day 501
Nose
SM009
Case 11
Day 504
Blood
SM009
Case 12
Day 507
Blood
SM009
Carrier 8
Day 557
Anus
SM009
Carrier 9
Day 561
Anus
SMV1
Carrier 10
Day 563
Anus
SMV1
Carrier 11
Day 585
Anus
SM009
Outbreak period 2 Case 10c
all infants whose clinical data have been used in this study.
‘Day 0’ indicates the day when the first epidemic BSI was detected Strains in bold type are blood isolates of SM009 (outbreak) pulsotype Strains in italic type are sporadic blood isolates a
Neonates with BSI and central nervous system infection
b
This strain has been isolated from the cerebrospinal fluid of Case 5
c
This neonate has been admitted from another neonatal unit of Budapest, and blood culture was positive in 24 h from admission
in five pediatric centers in Budapest, Hungary, has been approved by the Scientific and Research Ethics Council of the Hungarian Medical Research Council under the reference number 845-0/2010-1018EKU (39/PI/010). Clinical data collection during this outbreak analysis has been conducted as part of this multicenter study. Informed consent has been attained from the parents of
From May to June 2010 and August to September 2011 we investigated two outbreaks in a pediatric department. In all, we detected 12 neonates with nosocomial S. marcescens BSIs, and two blood isolates were detected between the two outbreak periods as well. In addition, 13 isolates causing colonization were investigated. Two S. marcescens strains isolated before the outbreak periods were also collected, aiming to compare these strains with the strains isolated during the outbreaks. Table 1 lists the investigated S. marcescens strains. During the outbreak period 1 nine clinical infections and six patients colonized with S. marcescens have been detected in the NICU from the first days of May until the middle of June 2010. The BSIs have been identified in two consecutive clusters. In the first cluster four premature neonates developed BSI at the NICU on the ‘Day 0’ and ‘Day 4’ of the outbreak, as shown in Table 1 (Cases 1–4). In the second cluster that appeared in early June (between Day 33 and Day 38 of the outbreak), five neonates had positive blood cultures (Cases 5–9). In case of one neonate (Case 5), the same strain could be detected from the cerebrospinal fluid, as well. During this second cluster infections were found not only at the NICU, but in the PICU, NSU and ISCU as well. During outbreak period 1 colonization could not be demonstrated in neonates that developed BSI. However, five other asymptomatic infants have been identified as a carrier of the pathogen (Carriers 1–5). These isolates originated from the nose, throat, trachea, anus and central venous catheter tip sample, as shown in Fig. 1. Between June 2010 to August 2011 two infants developed S. marcescens BSI at NSU sporadically (indicated as Sporadic 1 and Sporadic 2 in Table 1), and two cases of colonization have been found. In November 2010 one strain was isolated from an environmental sample of the ISCU, as described later. The outbreak strain was isolated from one infant (Carrier 5) two separate times: on Day 41 and on Day 164 as well. Outbreak period 2 started at the end of August 2011 (Day 487 in Table 1). Three cases with BSIs and two infants with colonization have been found in the NICU and in the PICU (Cases 10–12 in Table 1). The first neonate with BSI during this period (Case 10) was admitted from a neonatal ward of another hospital in Budapest, and blood culture became positive within 24 h from admission. One infant had eyecolonization besides the positive blood culture.
123
B. Iva´dy et al.
Fig. 1 Chronology of Serratia marcescens infections and colonization during the investigated period. This figure demonstrates that the epidemic PFGE genotype (SM009) could be detected during the whole study period. During the outbreaks SM009 caused all
bloodstream infections (black boxes). Between the outbreaks two patients was found to be colonized with the SM009 strain, and the SM009 type could be isolated from one water tap sample culture as well
After the second outbreak four children have been detected to be colonized with S. marcescens in November and December 2011. However, until the end of the investigated period, no more invasive infections have been observed.
solutions, sinks, ventilator and gas delivery devices, incubators) have been periodically cultured during and between the outbreaks. Hand cultures of HCWs were also obtained. During the outbreak periods altogether 326 environmental specimens were cultured. The only positive environmental culture was an epidemic S. marcescens strain isolated from a water tap in the ISCU between the outbreak periods, in November 2010 (indicated as ‘Environment’ on ‘Day 201’ in Table 1).
Outbreak control measures In May 2010 infection control staff including neonatologists, infectious diseases specialist, microbiologist, epidemiologist and infection control nurse started interventions to stop the outbreak. Education of hand hygiene guidelines has been started for all HCWs of the department; more attention was kept on cohort care and disinfection of environment. Implementation and adherence to the guidelines were regularly re-evaluated. As part of the intervention infected and colonized patients were transferred to an isolated ward, with dedicated nursing team and cleaning staff. Environmental culture survey Besides routine surveillance cultures from patients, various environmental samples (breast milk, formula, injection drugs, infusions, components of parenteral nutrition, disinfection
123
Clinical infections: presentation and outcome Clinical data about 12 S. marcescens BSIs from the investigated outbreak periods and two BSIs detected between the outbreak periods were collected. Data are summarized in Table 2. The babies’ age at the time of the positive blood culture was in the range of 5–165 days (median 37 days). The time from admission to the pediatric department until the development of infection was 1–110 days (median 24 days). The observed neonates were born at a median gestational age of 33.5 weeks (range 24–37 weeks) and the median birth weight was 2,070 g (range 600–3,450 g). Five (36 %) of all babies had an extremely low birth weight (ELBW). Three of the 14 infants (21 %) developed neuroinfection during the course of S. marcescens sepsis; the microbe could be isolated
NICU
NICU
NICU
NICU
Infant care unit
PICU
Infant care unit/PICU Infant care unit
Case 2
Case 3
Case 4
Case 5
Case 6
Case 7
Case 8
Neonatal surgery/PICU
Sporadic 2
NICU
NICU
Case 11
Case 12
83
43
5
59
110
11
92
51
165
25
5
21
31
10
Age (days)
25
25
37
34
33
35
36
37
34
33
24
24
31
34
Gestational age (weeks)
670
630
2,280
1,990
2,180
2,800
2,150
3,450
3,220
2,180
990
600
990
1,900
Birth weight (g)
Prematurity, BPD, ROP
Prematurity, IRDS
IUGR
Prematurity, tracheoesophageal fistula, pneumonia
Prematurity, jejunal atresia
Prematurity, dehydration
Smith–Lemli–Opitz syndrome
Undetermined CNS disease
Intestinal lymphangiectasia, otitis media
Prematurity, IRDS
Prematurity, IUGR, IRDS, NEC, PDA, IVH
Prematurity, IUGR, IRDS, NEC, PDA
Prematurity, IUGR, IRDS, NEC, PDA
Prematurity, IRDS
Underlying condition
18
43
1
54
110
3
32
4
2
24
24
24
31
10
In hospital days to BSI
Blood
Blood
Blood ? eye
Blood ? trachea
Blood
Blood
Blood
Blood
Blood
Blood and CSF
Blood
Blood
Blood
Blood and CSF
Cultures of S. marcescens
Sepsis
Sepsis
Sepsis
Sepsis
Sepsis
Bacteremia
Sepsis
Sepsis
Sepsis
Sepsis, meningitis
Sepsis
Sepsis
Sepsis, meningitis
Sepsis, meningitis, brain abscess
Clinical manifestation
Alive
Alive
Alive
Died
Died
Alive
Alive
Alive
Alive
Died
Died
Alive
Alive
Alive
28-day outcome
SM009
SM009
SM009
SM015
V1a
SM009
SM009
SM009
SM009
SM009
SM009
SM009
SM009
SM009
PFGE type
NICU Neonatal Intensive Care Unit, PICU Pediatric Intensive Care Unit, IRDS infantile respiratory distress syndrome, IUGR intrauterine growth retardation, NEC necrotizing enterocolitis, PDA patent ductus arteriosus, CNS central nervous system, BPD bronchopulmonary dysplasia, ROP retinopathy of prematurity
PICU
Case 10
Outbreak period 2
Neonatal surgery/PICU
Sporadic 1
Case 9
NICU
Hospital unit
Case 1
Outbreak period 1
Patient
Table 2 Epidemiological and clinical data of neonates with S. marcescens bloodstream infection (BSI)
Recurrent outbreaks of S. marcescens among neonates and infants
123
B. Iva´dy et al.
from the cerebrospinal fluid in two patients (both patients had positive blood cultures as well). Among the neonates with neuroinfection only one was an ELBW premature. Among low birth weight and premature neonates the underlying conditions were mostly IRDS, IUGR, PDA. Three out of four babies that died had more severe conditions and/or congenital malformations (intraventricular hemorrhage, jejunal atresia, and tracheoesophageal fistula). The 28-day mortality was found to be 29 % (4 of 14 cases), four of the five ELBW infants recovered from the S. marcescens BSI. Antimicrobial resistance All invasive isolates had the same antimicrobial resistance pattern. The strains causing BSIs were resistant to amoxicillin, amoxicillin/clavulanic acid, cefuroxime, and showed in vitro sensitivity to ceftriaxone, ceftazidime, cefepime, ciprofloxacin, trimethoprim/sulfamethoxazol, aminoglycosides (tobramycin, gentamicin, amikacin), pipercillin/tazobactam, imipenem and meropenem. Among strains that caused colonization, one isolate from the anus of a neonate was found to be resistant against ceftriaxone, piperacillin/tazobactam, carbapanems, aminoglycosides, and was sensitive only to ceftazidime, ciprofloxacin and levofloxacin. This strain had different PFGE type than any other strains isolated in the investigated period (see below). Molecular (PFGE) analysis PFGE analysis was performed for 16 BSI isolates, for one CSF isolate (from one patient that had a blood culture isolate as well), for 12 colonizing strains, one strain isolated from a water tap, and for two strains obtained from sporadic BSIs detected 6 months before the outbreak (these strains were retrospectively analyzed). Table 1 lists the S. marcescens isolates that have undergone PFGE analysis. In all, four different PFGE patterns have been found in our study: the SM009, SM010, SM015, SMV1/V1a. In our observation all blood culture strains isolated during the two outbreaks belonged to the SM009 PFGE pattern. The blood isolates detected between the outbreak periods represented two different PFGE types. Colonizing strains in this period belonged to the SM009 pulsotype and the strain isolated from a water tap between the outbreak periods proved to be SM009 as well, molecular investigations revealed that the strain responsible for the outbreaks (SM009) could be detected before, between and after the outbreaks at the pediatric department (Fig. 1). Discussion Serratia marcescens, a Gram-negative bacterium, is a wellknown pathogen of nosocomial outbreaks, especially in ICUs. Epidemiological, microbiological and clinical
123
characteristics of outbreaks in NICUs and PICUs have been intensively studied in the last decade. We report two consecutive outbreaks of S. marcescens BSIs in neonates and infants in three different wards of a pediatric center. We identified four different PFGE genotypes from which one was responsible for all BSIs during the two outbreak periods, but not for the sporadic BSIs between the outbreaks. The source of the infections could be identified with environmental cultures in several outbreaks reported recently. S. marcescens strains could be isolated from the hands of HCWs [5, 9, 15–17], hand washing soaps [15, 18, 19], sink drain [20], incubator [5], laryngoscope [4, 21], different drugs [22], parenteral nutrition [23], and baby shampoo [24]. During the outbreaks we could not detect the epidemic S. marcescens strain (PFGE type SM009) in the environment of the patients. However, this strain was isolated as a colonizing agent in asymptomatic patients during the outbreak periods, and detected in one culture of a water tap sample between the two outbreaks. We suppose that besides an environmental focus, colonized patients may serve as reservoirs and source for recurrent infections among NICU patients, as described earlier by many authors [25–28]. In our view, regular screening for specific pathogens to detect colonization is of particular importance in NICUs. For example, weekly screening for Gram-negative pathogens can help empiric antibiotic therapy [29] and identify specific pathogens (including Serratia) that are able to cause epidemics and recurrent infections in these settings. In our clinic’s NICU weekly screening cultures of anal specimens for multidrug resistant bacteria and ESBLproducing Gram-negatives have been performed regularly in all patients since 2008. In case of an outbreak or cluster of infections screening policy is extended to the causative organism (e.g., vancomycin-resistant enterococci or methicillin-resistant Staphylococcus aureus). During the periods of the described S. marcescens outbreaks in 2010–2011, surveillance has been extended to this pathogen, and altogether 3,386 screening cultures has been performed in these time periods. Results showed that during the whole investigated period there were almost always neonates in the NICU colonized with SM009 strain. However, only one environmental sample proved the presence of the pathogen, which was isolated in a period when no invasive infections were caused by this strain. Importantly, colonization with SM009 disappeared in January 2011, supposedly due to the infection control measures that stopped the transmission of the pathogenic strain. It is important that the first neonate with BSI in the outbreak period 2 was admitted from another hospital, and the blood culture was positive within 24 h from admission, suggesting that the SM009 was probably re-introduced to the pediatric clinic from the referring hospital.
Recurrent outbreaks of S. marcescens among neonates and infants
It would be useful to conduct epidemiological studies in other hospital settings in the same region to detect the predominant strains or clones that spread from hospital to hospital and ward to ward. Isolation of colonized and/or infected patients and barrier precautions are needed to control the horizontal spread of pathogens among patients in the same setting. During the outbreak periods in our institution, neonates colonized or infected by S. marcescens were isolated in a single box, managed by dedicated nurses, and using gloves, gowns and masks was obligatory for those who were caring for these patients. When a colonized patient was transferred to another ward, isolation and barrier precautions were strictly continued. Importantly, asymptomatic colonized patients were not treated with antibiotics. Recognizing that physicians may be an epidemiological link between the different wards, physicians were educated about nosocomial transmission of pathogens, and regular hand-hygiene educations were performed for all HCWs. In our observation, during outbreak period 1, the rate of neuroinfections (i.e., meningitis or brain abscess) as complication or focus of septicemia was high (33.3 %). Two neonates had meningitis, and one developed brain abscess, which is a known, but rare entity, as described by Hirooka et al. and Messerschmidt et al. [30, 31]. The rate of neuroinfections during neonatal S. marcescens outbreaks is 0–60 %. In a systematic analysis that reviewed 27 outbreaks involving 239 cases the rate was reported to be 7 % [2]. The highest rate was observed by Berger et al. [32], who described five cases of S. marcescens infections, from which three developed meningoencephalitis with brain abscess. Madani et al. [24] reported that the rate of neuroinfections during an outbreak in an NICU was 14 % (2 meningitis of 14 infected patients). Bizzarro et al. [33] analyzed sporadic neonatal S. marcescens infections compared to E. coli infections, and the rate of neuroinfections was found to be 24 and 7 %, respectively. The high rate of CNS involvement necessitates early treatment with antibiotics that are able to cross the blood–brain barrier in all neonates with S. marcescens infections, and early diagnostic measures (lumbar puncture, neuroimaging studies) to detect meningitis and/or brain abscess.
Conclusions Serratia marcescens can cause severe nosocomial outbreaks in neonates and infants. PFGE is a useful method to identify the predominant strain that may cause recurrent outbreaks. The outbreak strain can be detected in asymptomatic patients and in the environment without causing invasive disease in that period of time, but it can be the source of outbreaks later. Regular surveillance
cultures are needed even when the outbreak is controlled by isolation of infected patients, strict hand hygiene policies and continuous education. Early antibiotic therapy is needed in all cases of neonatal S. marcescens infections, always considering the possibility of CNS involvement. Acknowledgments We are very grateful to Natasa Pesti, Kla´ra To´th and Ma´ria Sze´na´si for their excellent assistance. The work was supported by Hungarian National Scientific Research Fund: OTKA108481. Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
References 1. Hejazi A, Falkiner FR. Serratia marcescens. J Med Microbiol. 1997;46(11):903–12. 2. Voelz A, Muller A, Gillen J, Le C, Dresbach T, Engelhart S, et al. Outbreaks of Serratia marcescens in neonatal and pediatric intensive care units: clinical aspects, risk factors and management. Int J Hyg Environ Health. 2010;213(2):79–87. doi:10.1016/ j.ijheh.2009.09.003. 3. Cimolai N, Trombley C, Wensley D, LeBlanc J. Heterogeneous Serratia marcescens genotypes from a nosocomial pediatric outbreak. Chest. 1997;111(1):194–7. 4. Jones BL, Gorman LJ, Simpson J, Curran ET, McNamee S, Lucas C, et al. An outbreak of Serratia marcescens in two neonatal intensive care units. J Hosp Infect. 2000;46(4):314–9. doi:10. 1053/jhin.2000.0837. 5. Jang TN, Fung CP, Yang TL, Shen SH, Huang CS, Lee SH. Use of pulsed-field gel electrophoresis to investigate an outbreak of Serratia marcescens infection in a neonatal intensive care unit. J Hosp Infect. 2001;48(1):13–9. doi:10.1053/jhin.2001.0947. 6. Villa J, Alba C, Barrado L, Sanz F, Del Castillo EG, Viedma E, et al. Long-term evolution of multiple outbreaks of Serratia marcescens bacteremia in a neonatal intensive care unit. Pediatr Infect Dis J. 2012;31(12):1298–300. doi:10.1097/INF. 0b013e318267f441. 7. Lai KK, Baker SP, Fontecchio SA. Rapid eradication of a cluster of Serratia marcescens in a neonatal intensive care unit: use of epidemiologic chromosome profiling by pulsed-field gel electrophoresis. Infect Control Hosp Epidemiol. 2004;25(9):730–4. doi:10.1086/502468. 8. Maragakis LL, Winkler A, Tucker MG, Cosgrove SE, Ross T, Lawson E, et al. Outbreak of multidrug-resistant Serratia marcescens infection in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2008;29(5):418–23. doi:10.1086/587969. 9. Friedman ND, Kotsanas D, Brett J, Billah B, Korman TM. Investigation of an outbreak of Serratia marcescens in a neonatal unit via a case-control study and molecular typing. Am J Infect Control. 2008;36(1):22–8. doi:10.1016/j.ajic.2006.12.012. 10. Al Jarousha AM. El Qouqa IA, El Jadba AH, Al Afifi AS. An outbreak of Serratia marcescens septicaemia in neonatal intensive care unit in Gaza City, Palestine. J Hosp Infect. 2008;70(2):119–26. doi:10.1016/j.jhin.2008.06.028. 11. Buffet-Bataillon S, Rabier V, Betremieux P, Beuchee A, Bauer M, Pladys P, et al. Outbreak of Serratia marcescens in a neonatal intensive care unit: contaminated unmedicated liquid soap and risk factors. J Hosp Infect. 2009;72(1):17–22. doi:10.1016/j.jhin. 2009.01.010.
123
B. Iva´dy et al. 12. Iosifidis E, Farmaki E, Nedelkopoulou N, Tsivitanidou M, Kaperoni M, Pentsoglou V, et al. Outbreak of bloodstream infections because of Serratia marcescens in a pediatric department. Am J Infect Control. 2011;. doi:10.1016/j.ajic.2011.03.020. 13. CDC. Standardized molecular subtyping of foodborne bacterial pathogens by pulsed-field gel electrophoresis: training manual. (Centers for disease control and prevention, Atlanta, GA). 2000. 14. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol. 1995;33(9):2233–9. 15. Manning ML, Archibald LK, Bell LM, Banerjee SN, Jarvis WR. Serratia marcescens transmission in a pediatric intensive care unit: a multifactorial occurrence. Am J Infect Control. 2001;29(2):115–9. doi:10.1067/mic.2001.114222. 16. Milisavljevic V, Wu F, Larson E, Rubenstein D, Ross B, Drusin LM, et al. Molecular epidemiology of Serratia marcescens outbreaks in two neonatal intensive care units. Infect Control Hosp Epidemiol. 2004;25(9):719–21. doi:10.1086/502466. 17. Villari P, Crispino M, Salvadori A, Scarcella A. Molecular epidemiology of an outbreak of Serratia marcescens in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2001;22(10):630–4. doi:10.1086/501834. 18. Rabier V, Bataillon S, Jolivet-Gougeon A, Chapplain JM, Beuchee A, Betremieux P. Hand washing soap as a source of neonatal Serratia marcescens outbreak. Acta Paediatr. 2008;97(10):1381–5. doi:10.1111/j.1651-2227.2008.00953.x. 19. Polilli E, Parruti G, Fazii P, D’Antonio D, Palmieri D, D’Incecco C et al. Rapidly controlled outbreak of Serratia marcescens infection/colonisations in a neonatal intensive care unit, Pescara General Hospital, Pescara, Italy, April 2011. Euro Surveill. 2011;16(24):16–18. 20. Maltezou HC, Tryfinopoulou K, Katerelos P, Ftika L, Pappa O, Tseroni M, et al. Consecutive Serratia marcescens multiclone outbreaks in a neonatal intensive care unit. Am J Infect Control. 2012;40(7):637–42. doi:10.1016/j.ajic.2011.08.019. 21. Cullen MM, Trail A, Robinson M, Keaney M, Chadwick PR. Serratia marcescens outbreak in a neonatal intensive care unit prompting review of decontamination of laryngoscopes. J Hosp Infect. 2005;59(1):68–70. doi:10.1016/j.jhin.2004.08.003. 22. Fleisch F, Zimmermann-Baer U, Zbinden R, Bischoff G, Arlettaz R, Waldvogel K, et al. Three consecutive outbreaks of Serratia marcescens in a neonatal intensive care unit. Clin Infect Dis. 2002;34(6):767–73. doi:10.1086/339046. 23. Arslan U, Erayman I, Kirdar S, Yuksekkaya S, Cimen O, Tuncer I, et al. Serratia marcescens sepsis outbreak in a neonatal
123
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
intensive care unit. Pediatr Int. 2010;52(2):208–12. doi:10.1111/j. 1442-200X.2009.02934.x. Madani TA, Alsaedi S, James L, Eldeek BS, Jiman-Fatani AA, Alawi MM, et al. Serratia marcescens-contaminated baby shampoo causing an outbreak among newborns at King Abdulaziz University Hospital, Jeddah Saudi Arabia. J Hosp Infect. 2011;78(1):16–9. doi:10.1016/j.jhin.2010.12.017. Sarvikivi E, Lyytikainen O, Salmenlinna S, Vuopio-Varkila J, Luukkainen P, Tarkka E, et al. Clustering of Serratia marcescens infections in a neonatal intensive care unit. Infect Control Hosp Epidemiol. 2004;25(9):723–9. doi:10.1086/502467. Miranda-Novales G, Leanos-Miranda B, Diaz-Ramos R, Gonzalez-Tejeda L, Peregrino-Bejarano L, Villegas-Silva R, et al. An outbreak due to Serratia marcescens in a neonatal intensive care unit typed by 2-day pulsed field gel electrophoresis protocol. Arch Med Res. 2003;34(3):237–41. David MD, Weller TM, Lambert P, Fraise AP. An outbreak of Serratia marcescens on the neonatal unit: a tale of two clones. J Hosp Infect. 2006;63(1):27–33. doi:10.1016/j.jhin.2005.11.006. Steppberger K, Walter S, Claros MC, Spencker FB, Kiess W, Rodloff AC, et al. Nosocomial neonatal outbreak of Serratia marcescens––analysis of pathogens by pulsed field gel electrophoresis and polymerase chain reaction. Infection. 2002;30(5):277–81. doi:10.1007/s15010-002-2141-y. Simon A, Tenenbaum T. Surveillance of multidrug-resistant gram-negative pathogens in high-risk neonates—does it make a difference? Pediatr Infect Dis J. 2013;32(4):407–9. Hirooka TM, Fontes RB, Diniz EM, Pinto FC, Matushita H. Cerebral abscess caused by Serratia marcescens in a premature neonate. Arq Neuropsiquiatr. 2007;65(4A):1018–21 S0004282X2007000600021. Messerschmidt A, Prayer D, Olischar M, Pollak A, Birnbacher R. Brain abscesses after Serratia marcescens infection on a neonatal intensive care unit: differences on serial imaging. Neuroradiology. 2004;46(2):148–52. doi:10.1007/s00234-003-1140-8. Berger A, Rohrmeister K, Haiden N, Assadian O, Kretzer V, Kohlhauser C. Serratia marcescens in the neonatal intensive care unit: re-emphasis of the potentially devastating sequelae. Wien Klin Wochenschr. 2002;114(23–24):1017–22. Bizzarro MJ, Dembry LM, Baltimore RS, Gallagher PG. Casecontrol analysis of endemic Serratia marcescens bacteremia in a neonatal intensive care unit. Arch Dis Child Fetal Neonatal Ed. 2007;92(2):120–6.
