An intermittent outbreak of Burkholderia cepacia contaminating hematopoietic stem cells resulting in infusate-related blood stream infections

Abstract

Microbial contamination of hematopoietic stem cells (HSC), used for autologous and allogenic transplantations, is rare but could cause serious blood stream infection in transplanted patients. These infections occur immediately, or later following the formation of biofilm on the catheter lumen. The present study describes an intermittent B. cepacia HSC contamination associated with nosocomial bacteremia: from October 2011 to April 2015, 17 B. cepacia strains were isolated in HSC bags (n = 14) and blood cultures (n = 3) in patients hospitalized in the National Bone Marrow Transplant Center. Two epidemiologic investigations in the National Blood Transfusion Center, allowing the isolation of three strains in hygiene samples, and four interventions in this institution were done. To identify the source of this contamination, a molecular investigation was done on 23 B. cepacia strains isolated in our center from 2007 to 2015. PFGE analysis revealed five clusters. The major cluster included 18 strains isolated from HSC bags (n = 14), blood culture (n = 1), and water cans and bath (n = 3). The second cluster (B) including only two and the remaining clusters (C, D, and E) contained single strains isolated before the epidemic period. These findings confirmed that the origin of the outbreak was the contaminated water used in the water bath during the thawing step of HSC bags. Based on this result, new sterile water was used for every defrosting, but HSC bags contamination persisted. In May 2015, the water bath was replaced with a dry bath and no B. cepacia strain was isolated from that date to April 2020.

Introduction

Burkholderia cepacia (B. cepacia) is a non-fermenting Gram-negative bacterium, previously known as Pseudomonas cepacia. This organism can survive in environment such as waste waters, soil, and any moist environment for a long time (Brooks et al., 2019). It is also an opportunistic pathogen, which causes different types of infections especially in patients with cystis fibrosis and immunocompromised patient (Avgeri et al., 2009). In onco-hematology units, outbreaks due to this nosocomial pathogen are frequent and are usually due to single contaminated source such as antiseptic solutions, intravenous solutions, nebulizer solutions, and medical devices (Abdallahet al., 2018). Antibiotic therapy for B. cepacia infection is a challenge, because this microorganism has intrinsic resistance to a lot of antibiotics (CA-SFM, 2012).

We describe the microbiological and molecular investigations into the first Tunisian intermittent outbreak of B. cepacia causing infusate-related blood stream infections arising from water-contaminating hematopoietic stem cells.

Material and methods

Clinical setting

The study was conducted in the National Bone Marrow Transplant Center (NBMTC) and the National Blood Transfusion Center (NBTC). The NBMTC is a university referral center specialized in all types of HSC transplantation and the treatment of patients with immunodeficiency in Tunisia. The treatment and storage of HSC from patient were done at the National Blood Transfusion Center (NBTC): After apheresis, the graft was analyzed and diluted or concentrated by centrifugation. To protect HSC, a cryoprotectant solution was added during the freeze step. All procedures were done under aseptic conditions. The HSC bags were conserved at −80°C.

During the entire process, quality control was done twice on the HSC bags with inoculation of samples into blood culture bottle.: Once before conservation and second after thawing in a water bath at 37°C. Blood cultures were taken from patients before and after transplantation.

Epidemiologic investigation

From October 2011 to May 2012, five B. cepacia strains were isolated in blood culture of transplanted patients (n = 2) and HSC bags (n = 3). An outbreak investigation was carried out. All hygiene samples taken (n = 20) in the area of the NBTC preparation of HSC (centrifuges, plasma extractor, water bath, etc.,) were negative for B. cepacia. Decontamination was done in the preparation area (Figure 1).

Figure 1.

Timeline of the outbreak investigation.

From July to December 2012, eight B. cepacia strains were isolated from HSC bags (n = 7) and blood culture of transplanted patients (n = 1). A second outbreak investigation was carried and five hygiene samples were done. B. cepacia was detected both in water bath (n = 2) and tap water cans (n = 1) which were used for transport of tap water to water bath. Based on these results, we suspected a contamination of the water cans and consequently the water bath. Therefore, we decontaminated the water bath, and sterile water was used for defrosting. No B. cepacia strain was detected in 2013 (Figure 1).

Four B. cepacia strains were detected in 2014 (n = 2) and 2015 (n = 2), these strains were isolated in HSC bags after the thawing step. In May 2015, the water bath was replaced with a dry bath and no B. cepacia strain was isolated from that date until now (Figure 1).It is worthy to note that there was no effect on the post-transplant course in patients receiving contaminated grafts.

Microbiologic investigation

The microbiological investigation included all B. cepacia strains, isolated from clinical or hygiene samples, from 2007 to 2015 strains (n = 23).

Inoculated blood culture bottles were incubated in the BacT/Alert 3D (Biomerieux) for 7 days. Hygiene samples were inoculated in trypticase soy agar for 3 days. Bacterial isolates from positive blood culture bottles and hygiene samples were identified by conventional methods (Gram staining and oxidase test) and Api20 NE system (Biomerieux®). Susceptibility testing was determined by disc diffusion method on Mueller-Hinton agar. The interpretation was done according to the recommendations of the CA-SFM. Isolates were stored in 10% glycerol at −80°C.

Molecular typing was done by Pulsed-field gel Electrophoresis (PFGE). The digestion was done with Xba1 enzyme, and restriction fragments were separated on a Chef DR-III (Bio-Rad) for 21H at 6 V/cm and 14°C (Coenye et al., 2002). The visualization was done under ultraviolet light (Gel Doc XR, Bio-Rad). The PFGE patterns were analyzed by the computer software Gelcompar II version 6.6 (Applied Math, Belgium).

Results

The PFGE analysis for the 23 strains revealed five clusters (Figure 2). The major cluster (A) included 18 strains isolated in blood culture of transplanted patient in the NBMTC (n = 1), HSC bags (n = 14), and hygiene samples (n = 3). These strains were isolated from October 2011 to April 2015. The second cluster (B) including only two strains was isolated in blood culture of transplanted patients in the NBMTC in May and July 2012 (Figure 2). The remaining clusters (C, D, and E) contained single strains isolated from 2007 to 2010.

Figure 2.

Dendrogram of PFGE fingerprinting of 23 B. cepacia strains after digestion with XbaI HSC: hematopoietic stem cells bags, Hyg: hygiene samples, Blood: blood culture Dice: (Opt 0.5%) (Tol 1%–1%).

All strains were susceptible to minocycline, levofloxacin, and co-trimoxazole and resistant to chloramphenicol and ticarcillin/clavulanic acid. Only the two strains, belonging to cluster B, were resistant to ceftazidime and meropenem.

Discussion

Our study investigated B. cepacia hematopoietic stem cells contamination associated to nosocomial bacteremia. Previous studies indicated that the contamination of HSC was rarely reported with a rate varying from 1.6% to 4.5% (Almeida et al., 2012). Moreover, coagulase-negative staphylococci were the most frequent species detected in HSC contamination and rarely Gram-negative strains, with a rate up to 50% and 4.7% of cases, respectively. The most frequent route of this contamination was by migration of coagulase-negative staphylococci from the skin, at the insertion site to the lumen of catheter (Almeida et al., 2012; Antoniewicz-Papis et al., 2020).

To the best of our knowledge, this is the first B. cepacia outbreak investigated and reported in Tunisia. Numerous cases of B. cepacia nosocomial infections have been reported all over the world. In India, for 4 years, at last four B. cepacia outbreaks were reported. Moreover, in the USA, at last 6 B. cepacia outbreaks were reported since 1991 (Abdallahet al., 2018).

The molecular typing showed five clusters with the predominance of cluster A which included 15 strains isolated from transplanted patients (n = 1) and HSC bags (n = 14), and three hygiene strains isolated from water bath (n = 2) and water cans (n = 1). Based on epidemiologic investigation and the results of PFGE, we conclude that HSC were contaminated during thawing in the water bath contaminated with B. cepacia. B. cepacia can remain viable for long time in water, liquids, and humid environment. Many B. cepacia outbreaks were due to the contamination of distilled water, saline flush syringes, and albuterol nebulization solution (Abdallahet al., 2018; Brooks et al., 2019). Moreover, several studies have reported B. cepacia outbreaks in intensive care units (ICUs) due to contaminated water. Such as in Indian hospitals, environmental sampling in ICUs showed that water was implicated as a major environmental source (Bhatia et al., 2017; Rastogi et al., 2019).

As to how B. cepacia got in the bag, we propose two hypotheses:

• - During the conservation period, micro-cracks have formed in the bag caused by the freezing and the condition of storage. On the thawing step and in the contaminated water bath, B. cepacia got in the bag through these micro-cracks.

• - If the organisms survived transfer on the bag, those administering may have failed to adequately disinfect the entry point of the administration set and thus administered the organism along with the HSC.

In the present study, we took one year to find the contamination source. Based on this result, new sterile water was used for every thawing, but HSC bags contamination persisted for more than three years, in spite of decontaminations, until the replacement of the water bath with a dry bath. Several studies reported outbreaks up to 4 years (Righi et al., 2013). These long periods are probably due to the difficulty to identify the real source of contamination and the resistance of B. cepacia to disinfectants and antiseptic solutions (Brooks et al., 2019; Righi et al., 2013).

All strains were susceptible to cotrimoxazole and two strains were resistant to ceftazidime and meropenem, which are the most common antibiotics used for the treatment of B. cepacia infections. Many studies reported up to 50% resistance to tested antibiotics and indicated that multi-drug resistant strains are common and are responsible for reduction of therapeutic options and high rates of mortality (Avgeri et al., 2009).

In our study, there was no effect on the post-transplant course in patients receiving contaminated grafts. Majado et al. reported no difference between contaminated and non-contaminated patients on the day the fever started, length of fever, blood transfusion requirements, and engraftment; however, length of hospitalization was significantly greater in patients receiving contaminated transplants (Majado et al., 2007).

This investigation has some limitations such as MLST typing which was not performed to determine the sequence type of the two clusters A and B. With this technique, we can compare our strains to isolates in other regions and publications. Moreover, no chemical and physical analyses of bags were done after a long period of conservation to confirm the presence of micro-cracks. These manipulations were not carried out due to the lack of materials.

In conclusion, we reported in this study for the first time, to the best of our knowledge, an intermittent outbreak of B. cepacia responsible of HSC contamination, due to HSC defrosting bags in contaminated water, associated to nosocomial bacteremia. This dissemination has been stopped after substitution of the water bath with a dry bath. Bacteriological control of HSC at the different processing steps is essential to reveal in time an eventual source of contamination to avoid the occurrence of outbreaks in immunocompromised patients and to preserve transplant prognosis.

ORCID iD

Anis Raddaoui https://orcid.org/0000-0003-0293-9735