Journal of Hospital Infection 98 (2018) 295e299

Available online at www.sciencedirect.com

Journal of Hospital Infection

journal homepage: www.elsevier.com/locate/jhin

Outbreak of Burkholderia cepacia pseudobacteraemia

caused by intrinsically contaminated commercial 0.5%
chlorhexidine solution in neonatal intensive care units
J.E. Song a, d, e, Y.G. Kwak a, d, *, T.H. Um b, C.R. Cho b, d, S. Kim b, I.S. Park b,
J.H. Hwang c, N. Kim c, d, G-B. Oh d
a
Department of Internal Medicine, Inje University Ilsan Paik Hospital, Goyang, Republic of Korea
b
Department of Laboratory Medicine, Inje University Ilsan Paik Hospital, Goyang, Republic of Korea
c
Department of Paediatrics, Inje University Ilsan Paik Hospital, Goyang, Republic of Korea
d
Infection Control Office, Inje University Ilsan Paik Hospital, Goyang, Republic of Korea
e
Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea

A R T I C L E I N F O S U M M A R Y

Article history: Background: Burkholderia cepacia is intrinsically resistant to certain antiseptics. The
Received 27 June 2017 authors noted a sudden increase in the frequency of isolation of B. cepacia from blood
Accepted 13 September 2017 cultures in a neonatal intensive care unit (NICU) of a university-affiliated hospital.
Available online 19 September Aim: To identify the source and intervene in the ongoing infections.
2017 Methods: The cases were defined as patients with positive blood cultures for B. cepacia in
an NICU between November 2014 and January 2015. Medical records were reviewed and
Keywords: NICU healthcare workers were interviewed. Samples of suspected antiseptics, blood cul-
Burkholderia cepacia ture bottles, cotton balls, gauze and a needle used in the NICU were analysed
Pseudobacteraemia microbiologically.
Chlorhexidine Findings: During the outbreak period, B. cepacia was identified in 25 blood cultures ob-
tained from 21 patients. The clinical features of the patients were suggestive of pseu-
dobacteraemia. Regarding environmental samples, B. cepacia was cultured from 0.5%
chlorhexidine gluconate (CHG) solution products that had been used as a skin antiseptic
during blood drawing in the NICU. The clinical B. cepacia isolate and two strains obtained
from 0.5% CHG exhibited identical pulsed-field gel electrophoresis patterns. After the CHG
products were withdrawn, the outbreak was resolved.
Conclusions: The pseudobacteraemia cases were caused by contaminated 0.5% CHG
produced by a single manufacturer. Stricter government regulation is needed to prevent
contamination of disinfectants during manufacturing. In addition, microbial contamination
of antiseptics and disinfectants should be suspected when a B. cepacia outbreak occurs in
hospitalized patients.
ª 2017 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Address: Department of Internal Medicine, Inje University Ilsan Paik Hospital, Juhwa-ro 170, Ilsanseo-gu, Goyang-si,
Gyeonggi-do, 10380, Republic of Korea. Tel.: þ82 31 910 7926; fax: þ82 31 910 7219.
E-mail address: ygkwak@paik.ac.kr (Y.G. Kwak).

https://doi.org/10.1016/j.jhin.2017.09.012
0195-6701/ª 2017 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
296 J.E. Song et al. / Journal of Hospital Infection 98 (2018) 295e299
Introduction
Burkholderia cepacia is an aerobic Gram-negative bacillus
that is widely distributed in the environment [1]. It can survive
in aqueous environments in hospital settings, from which it may
infect immunocompromised hosts and cause hospital outbreaks
[2]. Between November 2014 and January 2015, a sudden in-
crease in the frequency of isolation of B. cepacia from blood
cultures was seen in the neonatal intensive care unit (NICU) of
a university-affiliated hospital. After conducting surveillance
to find the source of the outbreak, the cause was identified and
the outbreak was terminated.
Methods
Setting
This study was conducted in a 600-bed university-affiliated
teaching hospital in Korea. The hospital has three intensive
care units, including a 30-bed NICU. The average number of pa-
tients hospitalized per month in the NICU in 2014 was 35. In early
January 2015, the infection control team (ICT) of the hospital
reported that B. cepacia was being isolated more frequently from
the blood cultures of NICU patients than previously.
Epidemiological investigation
The ICT started an investigation immediately. They visited
the NICU, interviewed the medical staff, and monitored the
blood culture process. In addition, they reviewed the medical
records of the patients who had blood cultures positive for B.
cepacia from the first patient on 20th November 2014. They also
checked compliance with infection control guidelines, and
evaluated possible common sources of contamination, including
injectables, preservatives and intravascular catheters.
Microbiological investigation
Fourteen environmental samples were collected aseptically
for analysis, including five unused aerobic and anaerobic blood
culture bottles stored in the NICU, three samples of 0.5%
3
Number of blood cultures
0
31/10/2014 10/11/2014
Figure 1. Isolation of Burkholderia cepacia in the neonatal intensive care unit before and after the outbreak.
chlorhexidine gluconate (CHG) solution (0.5 mL, 1.0 mL and
2.0 mL) in sealed bottles that had been stored but not used in
the NICU, four cotton balls soaked with 0.5% CHG before and
after use, one gauze pad and a scalp needle. These materials
were inoculated on to blood culture bottles or blood agar
plates, and incubated at 37 C for up to five days.
Phenotypic identification and antimicrobial susceptibility
profiling of B. cepacia blood isolates were performed using the
VITEK 2 system (bioMérieux, Marcy l’Étoile, France). The
standard breakpoints of the minimum inhibitory concentra-
tions were interpreted according to the guidelines of the
Clinical and Laboratory Standards Institute. Pulsed-field gel
electrophoresis (PFGE) typing of DNA digested with Xba1 (Bio-
Rad, Hercules, CA, USA) was performed in accordance with
previously described methods [3]. Restriction fragments were
separated on a CHEF DR-II apparatus (Bio-Rad) for 21 h at 6 V/
cm and 14 C. Subsequently, the gels were stained with
ethidium bromide and photographed under ultraviolet light
(Gel Doc XR, Bio-Rad).
Results
Clinical features
From 20th November 2014 to 4th January 2015, B. cepacia
was isolated from 25 blood cultures of 21 patients (Figure 1).
The total number of patients hospitalized in the NICU from
whom blood cultures were obtained during this period was 79.
Before the outbreak, only one case of B. cepacia bacteraemia
had occurred in the NICU (in 2012) since the official opening of
the hospital in 2000. Blood cultures were performed on all
patients on admission to the NICU because neonatal sepsis can
be asymptomatic and fatal [4,5]. Patients at high risk of
infection underwent repeated blood culture tests at weekly
intervals. Table I shows the clinical characteristics of the pa-
tients with B. cepacia in blood cultures.
The median hospital stay until a positive blood culture was
four days (range 0e59). In five patients (23.8%), B. cepacia was
isolated from blood cultures taken on the day of admission.
Twenty (95.2%) of the 21 patients had no fever at the time of
blood culture collection. Although B. cepacia was isolated from
20/11/2014 30/11/2014 10/12/2014 20/12/2014 30/12/2014 09/01/2015 19/01/2015 29/01/2015
Date of blood culture
 J.E. Song et al. / Journal of Hospital Infection 98 (2018) 295e299 297

Table I
Clinical characteristics of patients with Burkholderia cepacia in blood cultures
Patient Date of birth Sex Weight Admission Diagnosis Blood culture Fever Ventilator Outcome
(g)a date positive date of care
B. cepacia
1 23/10/2014 M 2290 23/10/2014 Preterm, RDS 20/11/2014 No Yes Survived
2 04/12/2014 F 2940 04/12/2014 RDS 04/12/2014 No No Survived
11/12/2014
3 02/12/2014 M 1400 02/12/2014 Preterm, RDS 05/12/2014 No No Survived
4 17/11/2014 M 2460 17/11/2014 Preterm, RDS 06/12/2014 Yes Yes Survived
30/12/2014
03/01/2015
5 23/11/2014 M 2620 26/11/2014 Neonatal seizure, 06/12/2014 No No Survived
norovirus enteritis
6 27/11/2014 M 2370 27/11/2014 Preterm, RDS 08/12/2014 No Yes Survived
7 27/11/2014 F 2530 27/11/2014 Preterm, RDS 11/12/2014 No No Survived
8 03/12/2014 M 4280 20/12/2014 Sepsis, rotavirus enteritis 23/12/2014 No No Survived
9 13/11/2014 M 2440 13/11/2014 Preterm, RDS, sepsis due 24/12/2014 No Yes Survived
to Candida spp.
10 24/12/2014 M 3020 24/12/2014 Sepsis 24/12/2014 No Yes Survived
11 18/12/2014 M 2160 18/12/2014 Preterm, RDS 22/12/2014 No Yes Survived
12 22/12/2014 M 1470 22/12/2014 Preterm, RDS 25/12/2014 No Yes Survived
13 26/12/2014 M 1350 26/12/2014 Preterm, RDS, BPD 29/12/2014 No Yes Survived
14 29/12/2014 F 2340 29/12/2014 Preterm, RDS 29/12/2014 No No Survived
01/01/2015
15 29/12/2014 F 1200 29/12/2014 Preterm, RDS, BPD 29/12/2014 No No Survived
16 17/11/2014 M 1070 18/11/2014 Pericardial effusion 28/12/2014 No No Survived
17 26/12/2014 F 3760 27/12/2014 Second degree burn, RDS 29/12/2014 No Yes Survived
18 03/11/2014 F 1750 03/11/2014 Preterm, RDS, PDA 01/01/2015 No Yes Survived
19 01/01/2015 M 2240 01/01/2015 Preterm, RDS, PSVT 01/01/2015 No Yes Survived
20 21/12/2014 F 4020 21/12/2014 Sepsis 03/01/2015 No Yes Survived
21 29/12/2014 F 1280 29/12/2014 Preterm, RDS, BPD 04/01/2015 No No Survived
RDS, respiratory distress syndrome; BPD, bronchopulmonary dysplasia; PDA, patent ductus arteriosus; PSVT, supraventricular tachycardia; M, male;
F, female.
a
Weight at the time of the first blood culture positive for B. cepacia.

blood cultures, the attending neonatologist suspected
contamination or false-positive results rather than clinical
infection because there were no changes in vital signs, leuko-
cyte counts or C-reactive protein; as such, the attending
neonatologist did not change or add antibiotics to treat
B. cepacia. Based on these results, contamination of antiseptic
or a breakdown in the aseptic technique during blood culture
collection was suspected as a possible source of the outbreak.
Epidemiological investigation
Blood cultures were performed using the aseptic technique,
and it was identified that 0.5% CHG solution, rather than 10%
povidone-iodine, had been used as a skin disinfectant since
2012 when taking blood for cultures in the NICU. The 0.5% CHG
solution (Hexitan) used in the NICU was a commercial product
purchased by the hospital pharmacy, and it was not diluted in
the hospital. The 0.5% CHG solution was used in the NICU and
surgical intensive care units (SICU) alone; it was not used in
other wards in the hospital.
Use of disinfectants and antiseptics in the NICU was in
compliance with the infection control guidelines of the hospi-
tal. Sterile cotton balls were placed in a sterile container and
the antiseptics were poured aseptically. Antiseptic-soaked
cotton balls were used to disinfect the skin or mucus mem-
brane for up to 24 h; any unused cotton balls were discarded.
Microbiological investigation
The 26 B. cepacia isolates showed similar patterns of anti-
biotic susceptibility. Most of the B. cepacia isolates were sus-
ceptible to ceftazidime (24/25, 96.0%) and meropenem (20/25,
80.0%), but resistant to cefepime (21/25, 84.0%), tigecycline
(24/25, 96.0%), gentamicin (100%), imipenem (100%), piper-
acillin/tazobactam (100%) and ciprofloxacin (100%). No com-
mon sources, including intravenous medications, intravascular
catheters and invasive procedures, were identified among the
patients by the investigators.
Of the 14 environmental samples tested, B. cepacia was
detected in seven samples: three samples of 0.5% CHG solution
(0.5 mL, 1.0 mL and 2.0 mL) and four cotton balls soaked with
0.5% CHG. Other environmental samples did not grow bacteria.
Only one clinical isolate was collected from Patient 21 because
the other isolates had been discarded before the investigation.
The clonal identities of the single clinical isolate and two
environmental isolates (one of the three isolates from the 0.5%
CHG solution and one of the four isolates from cotton balls
soaked with 0.5% CHG) were compared. These clinical and
298 J.E. Song et al. / Journal of Hospital Infection 98 (2018) 295e299
environmental B. cepacia isolates were fingerprinted using
PFGE to investigate clonal identity; their PFGE patterns were
identical (Figure 2).
Management of the outbreak
When the ICT recognized the B. cepacia outbreak, they
checked and reminded medical staff in the NICU of the basic
principles of infection control, including hand hygiene and
disinfection of medical devices. After the 0.5% CHG solution
had been identified as the source of B. cepacia, its use in the
hospital was suspended (6th January 2015). All NICU staff were
re-educated to use 10% povidone-iodine alone for skin anti-
sepsis. After complete recall of the 0.5% CHG product, no
further B. cepacia were isolated from blood (Figure 1). The
date of the last isolation of B. cepacia from blood was 4th
January 2015. On reaching this conclusion, the Korean Ministry
of Food and Drug Safety recalled Hexitan 0.5% CHG
solution, Lot Nos D021, D022, D026 and D027, manufactured on
4th and 7th November 2014, because of bacterial contamination
1 2 3 M
Figure 2. Pulsed-field gel electrophoresis patterns of Burkholderia
cepacia isolates from clinical (1) and environmental samples (2, 3).
Lane 1, isolate from Patient 21; lane 2, isolate from 0.5% chlor-
hexidine gluconate (CHG) solution; lane 3, isolate from a cotton
ball soaked with 0.5% CHG; lane M, molecular weight standard.
(Medical Products Safety Division Document 12475). The CHG
product used in the NICU was Lot No. D021, but this was not
used in the SICU. The other recalled lots (D022, D026 and D027)
were not used in the study hospital.
Discussion
B. cepacia is rarely detected as a pathogen in nosocomial
infections [6]. Nevertheless, it is widely distributed in the
hospital environment and is a potential cause of hospital out-
breaks [7]. The detection of B. cepacia from clinical samples
usually suggests an environmental source. Nosocomial spread
of B. cepacia most frequently occurs because of contamination
of disinfectant solutions used to clean re-usable patient
equipment, such as nebulizers and pressure transducers [8,9].
Burkholderia spp. can remain viable for many months in water,
and pharmaceutical water can also be a source of these bac-
teria in industrial settings. Maintaining product sterility is a
problem faced by many manufacturing industries, and a recent
survey of the pharmaceutical industry indicated that B. cepacia
is one of the leading causes of pharmaceutical product recalls
[10]. In Korea, there are no regulations that require the ste-
rility of antiseptic and disinfectant products to be proven
routinely; such verification is left to the manufacturer’s
discretion [1].
CHG is a widely used antiseptic agent for skin disinfection,
handwashing and oral care [11]. It impairs the integrity of the
bacterial membrane, causing loss of periplasmic enzymes and
cytoplasmic components; at higher concentrations, it leads to
coagulation of bacterial cytoplasm [12]. Germicides may
become contaminated as a result of improper manufacturing
techniques, or during shipping (intrinsic contamination),
manipulation or use within a healthcare facility (extrinsic
contamination). In this case, the 0.5% CHG solution used in the
NICU had been intrinsically contaminated during manufacture.
During the seven-week outbreak, B. cepacia was isolated from
the blood cultures of 21 of 79 (26.6%) patients hospitalized in
the NICU. Most of these patients were premature infants or
high-risk neonates who needed ventilator support. It is difficult
to explain why B. cepacia was only isolated from some patients,
despite use of the same CHG lot in all hospitalized patients in
the NICU. The high rate of B. cepacia pseudobacteraemia in
such neonates may have been due to their thinner skin, which is
more susceptible to damage than that of other patients. In
addition, the difference in blood culture procedures among
healthcare practitioners and the degree of bacterial contami-
nation of the CHG used may have influenced the culture results.
Best practice regarding skin antisepsis in the NICU is unclear,
and the choice of antiseptics for neonates remains contentious
[13]. Most topical antiseptics can cause irritation and skin er-
ythema in neonates. Serious chemical skin burns have been
reported with the use of isopropyl alcohol, 0.5% CHG with 70%
alcohol, both alcohol- and aqueous-based 2% CHG, and iodine
solution [14]. CHG antiseptics are widely used in neonates, but
there are potential safety issues relating to their use in preterm
infants, such as possible toxic effects related to their increased
absorption, and immature organ function resulting in
decreased ability to clear the CHG. In the case of povidone-
iodine, preterm infants are at a higher risk for iodine over-
load due to decreased renal clearance and the inability to
regulate iodine uptake into the thyroid. Thus, many antiseptic
preparations in varying concentrations and combinations are
 J.E. Song et al. / Journal of Hospital Infection 98 (2018) 295e299
used for topical antisepsis in NICU. A survey of tertiary-level
NICUs found at least seven different preparations in use,
among which CHG predominated [14]. CHG was used at various
concentrations with or without isopropyl alcohol.
B. cepacia has caused outbreaks in intensive care units, the
sources of which include the respiratory route, nebulized
medication, mouthwash and povidone-iodine [15e19]. Re-
ported outbreaks of B. cepacia include both true bacteraemia
and pseudobacteraemia [2,20]. Pseudobacteraemia can be
caused by contamination during culture. Ko et al. reported
pseudobacteraemia associated with intrinsically contaminated
commercial 0.5% CHG solution, similar to the present study,
but they did not analyse clonal identity [1]. Gravel-Tropper
et al. reported 13 cases of pseudobacteraemia in the NICU,
and found that the blood gas analyser in the NICU was heavily
contaminated with B. cepacia [21]. In paediatric patients with
cystic fibrosis, B. cepacia may cause severe pneumonia, which
can increase the mortality rate [22]. Although the bacteria
differ slightly, Woo et al. reported a similar pseudobacter-
aemia outbreak in an emergency room caused by Pseudomonas
oryzihabitans from a contaminated saline gauze canister [23].
In the present case, it was fortunate that the organism did not
cause a true bacteraemia.
In conclusion, this article reports an outbreak of B. cepacia
pseudobacteraemia caused by intrinsically contaminated 0.5%
CHG solution from a single manufacturer. Stricter government
regulation is needed to prevent contamination of disinfectants
during manufacture. In addition, microbial contamination of
antiseptics and disinfectants should be suspected when a B.
cepacia outbreak occurs in hospitalized patients.
Acknowledgment
The authors wish to thank J.Y. Lee, Infection Control Office
of Asan Medical Center, Seoul for conducting PFGE.
Conflict of interest statement
None declared.
References
[1] Ko S, An HS, Bang JH, Park SW. An outbreak of Burkholderia
cepacia complex pseudobacteremia associated with intrinsically
contaminated commercial 0.5% chlorhexidine solution. Am J
Infect Control 2015;43:266e8.
[2] Heo ST, Kim SJ, Jeong YG, Bae IG, Jin JS, Lee JC. Hospital
outbreak of Burkholderia stabilis bacteraemia related to
contaminated chlorhexidine in haematological malignancy pa-
tients with indwelling catheters. J Hosp Infect 2008;70:241e5.
[3] Tseng SP, Tsai WC, Liang CY, Lin Y-S, Huang J-W, Chang C-Y, et al.
The contribution of antibiotic resistance mechanisms in clinical
Burkholderia cepacia complex isolates: an emphasis on efflux
pump activity. PLoS One 2014;9, e104986.
[4] Tewabe T, Mohammed S, Tilahun Y, Melaku B, Fenta M,
Dagnaw T, et al. Clinical outcome and risk factors of neonatal
sepsis among neonates in Fe lege Hiwot referral Hospital, Bahir
Dar, Amhara Regional State, North West Ethiopia 2016: a retro-
spective chart review. BMC Res Notes 2017;10:265.
[5] Berardi A, Buffagni AM, Rossi C, Vaccina E, Cattelani C,
Gambini L, et al. Serial physical examinations, a simple and
299
reliable tool for managing neonates at risk for early-onset sepsis.
World J Clin Pediatr 2016;5:358e64.
[6] Martin M, Christiansen B, Caspari G, Hogardt M, von Thomsen AJ,
Ott E, et al. Hospital-wide outbreak of Burkholderia contaminans
caused by prefabricated moist washcloths. J Hosp Infect
2011;77:267e70.
[7] Lee S, Han SW, Kim G, Song DY, Lee JC, Kwon KT. An outbreak of
Burkholderia cenocepacia associated with contaminated chlor-
hexidine solutions prepared in the hospital. Am J Infect Control
2013;41:e93e6.
[8] Kaitwatcharachai C, Silpapojakul K, Jitsurong S, Kalnauwakul S.
An outbreak of Burkholderia cepacia bacteremia in hemodialysis
patients: an epidemiologic and molecular study. Am J Kidney Dis
2000;36:199e204.
[9] Ghazal SS, Al-Mudaimeegh K, Al Fakihi EM, Asery AT. Outbreak of
Burkholderia cepacia bacteremia in immunocompetent children
caused by contaminated nebulized sulbutamol in Saudi Arabia.
Am J Infect Control 2006;34:394e8.
[10] Jimenez L. Microbial diversity in pharmaceutical product recalls
and environments. PDA J Pharm Sci Technol 2007;61:383e99.
[11] McDonnell G, Russell AD. Antiseptics and disinfectants: activity,
action, and resistance. Clin Microbiol Rev 1999;12:147e79.
[12] Coenye T, Van Acker H, Peeters E, Sass A, Buroni S, Riccardi G,
et al. Molecular mechanisms of chlorhexidine tolerance in Bur-
kholderia cenocepacia biofilms. Antimicrob Agents Chemother
2011;55:1912e9.
[13] Heron TJ, Faraday CM, Clarke P. The hidden harms of Matching
Michigan. Arch Dis Child Fetal Neonatal Ed 2013;98:F466e7.
[14] Ponnusamy V, Venkatesh V, Clarke P. Skin antisepsis in the
neonate: what should we use? Curr Opin Infect Dis
2014;27:244e50.
[15] Conly JM, Klass L, Larson L, Kennedy J, Low DE, Harding GK.
Pseudomonas cepacia colonization and infection in intensive care
units. CMAJ 1986;134:363e6.
[16] Weems Jr JJ. Nosocomial outbreak of Pseudomonas cepacia
associated with contamination of reusable electronic ventilator
temperature probes. Infect Control Hosp Epidemiol
1993;14:583e6.
[17] Hamill RJ, Houston ED, Georghiou PR, Wright CE, Koza MA,
Cadle RM, et al. An outbreak of Burkholderia (formerly Pseudo-
monas) cepacia respiratory tract colonization and infection
associated with nebulized albuterol therapy. Ann Intern Med
1995;122:762e6.
[18] Matrician L, Ange G, Burns S, Fanning WL, Kioski C, Cage GD,
et al. Outbreak of nosocomial Burkholderia cepacia infection and
colonization associated with intrinsically contaminated mouth-
wash. Infect Control Hosp Epidemiol 2000;21:739e41.
[19] Panlilio AL, Beck-Sague CM, Siegel JD, Fanning WL, Kioski C,
Cage GD, et al. Infections and pseudoinfections due to povidone-
iodine solution contaminated with Pseudomonas cepacia. Clin
Infect Dis 1992;14:1078e83.
[20] Lee JK. Two outbreaks of Burkholderia cepacia nosocomial
infection in a neonatal intensive care unit. J Paediatr Child
Health 2008;44:62e6.
[21] Gravel-Tropper D, Sample ML, Oxley C, Toye B, Woods DE,
Garber GE. Three-year outbreak of pseudobacteremia with Bur-
kholderia cepacia traced to a contaminated blood gas analyzer.
Infect Control Hosp Epidemiol 1996;17:737e40.
[22] Speert DP, Henry D, Vandamme P, Corey M, Mahenthiralingam E.
Epidemiology of Burkholderia cepacia complex in patients with
cystic fibrosis, Canada. Emerg Infect Dis 2002;8:181e7.
[23] Woo KS, Choi JL, Kim BR, Kim JE, Kim KH, Kim JM, et al. Outbreak
of Pseudomonas oryzihabitans pseudobacteremia related to
contaminated equipment in an emergency room of a tertiary
hospital in Korea. Infect Chemother 2014;46:42e4.