An outbreak of nosocomial infectious bacteria on the surfaces of medical instruments and the wounds of patients shortly after admission to the newly built burn intensive care unit

Background & Aims: Nosocomial infection (NI) is a significant problem in burn intensive care units (BICUs). Although a number of strict infection control practices had been carried out for preventing NI bacterium infection, the spreads or even outbreaks of these pathogens were often reported. Here we monitored the infection in a newly built BICU and tried to discover the origin and transmission route of NI. Methods: Five burn patients were gradually transferred or admitted into a newly built BICU. The surfaces of medical instruments were cleaned with detergent and later with chlorine disinfectant . NI bacteria colonies on surfaces of medical instruments, gloves, and burn wounds were sampled using Replicate Organism Detection and Counting (RODAC) plates. Gram-positive coccus and Gram-negative bacilli were identified, followed by antimicrobial susceptibility tests for Staphylococcus aureus and Acinetobacter baumannii. The shift of these two bacteria was depicted with a heat-map. The phylogenetic grouping was analysed based on the Enterobacterial repetitive intergenic consensus-polymerase chain reaction (ERIC-PCR) fingerprint. Results: Colonies of S.aureus and A. baumannii were gradually discovered on several instruments during the first 5 days when the detergent was used. Conversely, since the sterilization, almost no S.aureus or A. baumannii colonies could be found. Furthermore, colonies were found on the nursing assistant’s gloves at day 0, which was attributed to her repeated gloves use after her touch of the body of the first admitted patient. Simultaneously, the wounds of the other four admitted patients were colonized with same strains of A. baumannii starting at day 1 or day 2. Conclusions: The NI bacteria rapidly colonized the surfaces of the medical instruments, which was facilitated by using detergent but was eliminated shortly after by using a disinfectant. The colonization of NI bacteria on the wounds of burn patients might have come from the nursing assistant’s gloves.

were symptomed by discharge from the wound, fever, cough, shortness breath, burning with urination or difficult urinating, headache, nausea, vomiting and diarrhea. These infections were related to person to person contact, like unclean hands, the contact of person to medical instruments such as catheters, respiratory machines, and the abuse use of antibiotics. Main types of NIs were bloodstream infection, urinary tract infection, surgical site infection, and ventilator associated pneumonia (VAP) [1].
The infection will induce the exaggerated immune response from the host characterized by an increased cytokines release, with multiple organs disfunction, like sepsis [2]. Nowadays, the preventive actions have been considered an effective way to control NI. Its inspections, including the determination and interference of the dissemination of the pathogen, have been reported as effective measures of reducing infection rates [1]. Protective isolations, such as hand washing, the use of gloves and non-permeable aprons, and isolation of infected patients are effective methods to prevent pathogen transmission among patients. Although lots of strict infection control practices had been carried out in order to prevent various NI bacterium, the spreads or even outbreaks of these pathogens in the hospital environment were often reported [3][4][5].
51% patients have been reported by Dasgupta, et al. to have NI in ICUs. The most frequently reported sites of acquired infections in ICUs were lungs (64%), abdominal (19%) and blood stream (15%) which added an additional cost of 3.5 billions USD per year [6].
In burn units, NI has also been a significant problem for decades. It has increased the duration of hospital stays by 18 days on average and has increased the cost of treatment by a mean of 502 US dollars in a burn unit in Turkey [7]. BICUs admit patients with moderate-to-severe burns, those in shock during the early stage of burn injuries, with aggressive infections that need to be isolated, and patients with severe inhalation injuries who need to receive tracheotomy or even a ventilator. The age, length of hospital stay, and total burn surface area of the patients with NI infection were higher than those non-infected patients in BICUs [8]. Although several care and isolation practices lead to the decrease of infection rates, no "gold standard" of protective strategies can be successfully used in every settings, which implied some other putative neglected factors might play important roles in the occurrence of NIs in BICUs [9].
For high-risk patients, relocation to a vacant ward is regarded as an effective measure to prevent the outbreak of a NI [10,11]. This finding may be due to fewer NI bacteria in the new environment in the hospital. If patients carrying NI bacteria were transferred into another department, an outbreak of NI might emerge. It has been observed that patients relocated within hospitals were responsible for spreading multidrug-resistant Klebsiella pneumoniae [12]. However, the origin and transmission route of this bacterium in a new setting in a hospital has not been fully understood. The aim of the study is to trace the transmission path of the NI bacteria in the new environment in the hospital and to determine an effective way to interrupt the transmission of NI bacteria to preserve a relatively sterile environment for patients in the BICU.

8.
Cluster analysis DNA fingerprint fragments were given a score of '1' (presence of observed bands) or a score of '0' (absence of observed bands). Dendrogram analysis was performed using the software package NTsyspc (Applied biostatistics Inc. Port Jefferson, NY) based on the score's matrix, with a 1% tolerance limit and Dice's similarity coefficient using unweighted pair-group arithmetic average clustering (UPGMA).
Clusters were identified according to an 80% similarity cut-off.

Results
Disinfectant rather than detergent prevented the transmission of NI bacteria in the environment after the relocation to the newly built BICU.
During the first 22 days after the relocation to the newly built BICU, the surfaces of the instruments in the new department were sampled using RODAC plates. All the colonies of Gram-positive coccus and Gram-negative bacilli on the plates were counted and identified. Twenty-three Gram-positive coccus or Gram-negative bacilli species were detected on the surfaces of the instruments in the department, with most of them having few colonies, except for S.aureus and A. baumannii. Therefore, we focused on the spread of these two high-density species and their drug-resistance. Before the burn patients were transferred into the new department from the old department (day 0), neither S.aureus nor A.
baumannii could be detected on the surfaces of instruments by utilizing the RODAC plates. From day 0 to day 5 after the first five patients were transported into the department, surfaces of the instruments were wiped with detergent twice a day. By applying the RODAC plates, colonies of S.aureus were gradually discovered on the surfaces of several items one hour after cleaning instruments, including the computer mouse of the CMS, a computer mouse on the desk of the nurse's workstation, a food conveyer, and the desktop at the nurse workstation and in the dressing room, during the first five days. Simultaneously, colonies of A. baumannii were found on the food conveyer, the computer mouse on the desk of the doctor workstation, the wall next to a patient's bedside, and the desktop in the disposal room, in the treatment room and at the doctor workstation. Starting from day 6, we carried out sterilization using chlorine disinfectant instead of cleaning using detergent, which led to the disappearance of S.aureus and A. baumannii from the surfaces of the instruments in the BICU (Fig 1).

Nosocomial infection bacteria on the surfaces of the instruments were quickly removed after the sterilization
To evaluate the clearance of the NI bacteria by using detergent compared with disinfectant, colonies of A. baumannii, which was the dominant NI bacteria, were observed before and after the cleaning and sterilization procedures were carried out (Fig 2). As a result, detergent could not reduce the colony amounts on the surfaces of the food conveyer and the desktop in the treatment room, in the dressing room, and in the disposal room from day 3 to day 5. Conversely, almost no colonies could be observed on the RODAC plates from any of the surfaces from day 6 to day 8 after the disinfectant was utilized.
The prevalence of A. baumannii was caused by the contamination of the nursing assistant's glove.
Since the first burn patient was transferred from the old ward into the new ward, the shift of the amount of the isolated colonies of A. baumannii on the wounds that were most likely to be touched of every one of the five patients was observed during the first 5 days. On day 0, the first patient, numbered as BI001, was colonized with A. baumannii on his burn wound. However, from day 1, isolates of A. baumannii were gradually discovered on the wounds of the other 4 patients as well.
Unfortunately, despite the use of antibiotics, A. baumannii could no longer be eliminated due to their multidrug-resistance. We simultaneously examined the colonies on gloves of three doctors, four nurses and the only nursing assistant in the BICU before they contacted these patients. As a result, no colony was discovered on doctors' and nurses' gloves, but colonies with a concentration of 142 cfu/100 cm 2 were found on the nursing assistant's gloves on day 0. Hence, we inquired with the nursing assistant and learned that she did not replace her gloves after she had touched the body of patient BI001. To inhibit the spread of NI bacteria, the usage of an exclusive pair of gloves before she touches every patient is imperative. However, though she replaced her gloves before she wiped the surfaces of the instruments on day 1, she did not replace her gloves after the cleaning procedure, and she still had a bacterial concentration of 125 cfu/100 cm 2 colonies on her gloves until she began to change her gloves after she cleaned the surfaces of instruments from day 2.
We examined the drug-resistance spectrums of A. baumannii from the burn wounds of the patients and the gloves of the nursing assistant. Furthermore, as the gloves of the nursing assistant had been replaced since day 2 before she touched the new patients, the drug-resistance of A. baumannii on the surfaces in the instruments could no longer be discovered. As a result, three different spectrums were discovered ( To further reveal the transmission pathway of A. baumannii, all colonies collected from burn wounds of the five patients, the nursing assistant's gloves, the surfaces of a food conveyer, the desktop in the disposal room, and the wall next to a patient's bedside were characterized by phylogenetic grouping based on ERIC-PCR fingerprint. All colonies on the same objects were grouped as the same cluster, except for two strains (AB001a_4d and AB001_4d) from BI004 at day 4. Therefore, we selected one isolate on the surface of every object and two isolates on the wounds of BI004, and we grouped them using ERIC-PCR. Twenty-two isolates were grouped into two clusters, while the remaining 2 isolates were unclustered. C1 was the dominant cluster, with 20 isolates. All isolates, including AB001_at day 0, from patient BI001 from day 1 to day 4 before and after being transferred to the new ward, were clustered together. Isolates from the nursing assistant's gloves on the first two days, the surfaces of a food conveyer at day 1 and other patients from day 1 to day 4 were clustered with those from patient BI001. Isolates from the surfaces on the desktop in the disposal room (DD001_1d) and wounds of patient BI004 (AB004_1d) at day 1 were grouped into cluster 2 (C2). The individual strain of WA001_1d from the surface of the wall next to a patient's bedside in the department, and another isolate from BI004 at day 4 (AB001a_4d), isolated from BI001, was the sole member in C3 or C4, respectively.

Discussion
The However, in the current study, we did not find any colonies on doctors' or nurses' gloves. Meanwhile, although the nursing assistant wore gloves in our BICU, she did not understand the importance of changing to a fresh pair of gloves before she contacted the body of the next patient. Based on the molecular typing analysis, the first patient (BI001) was discovered as being colonized by A. baumannii on his burn wound, which was in the same cluster (C1) with the strains isolated from the nursing assistant's gloves at day 1 and day 2 after the relocation. From then on, the colonies of A. baumannii were gradually isolated from the burn wounds of other patients. Based on the dendrogram analysis and the occurrence time of these strains, we deduced that the spread pathway of A. baumannii might be from patient BI001 via the nursing assistant's gloves to the other four patients shortly after the relocation. Patient BI001 harboured another strain of A. baumannii (AB001a_4d) on his burn wounds at day 4 according to the cluster analysis. This strain did not spread to other patients, possibly due to its relative low density, the utilization of new gloves before contacting the next patients from day 3, and its drug-sensitivity to most of the antibiotics. This finding also implied that two strains of A. Availability of data and materials

Ethics approval and consent to participate
The study was approved by the Ethics Committees of Wuxi No.3 people's hospital (IEC201607002). All participants were consented to participate.

Consent for publication
All participants have provided written consent to participate in this study that included for the publication.   Figure 1 Spread of S.aureus and A. baumannii on the surfaces of different instruments in the BICU was almost inhibited after the environment had been sterilized daily. The heat map shows the shift of colony counts of (a) S.aureus and (b) A. baumannii before (day 0) and after (day 1 to day 22) the relocation to the newly built BICU. From day 0 to day 5, detergent was used to clean the surfaces. From day 6 to day 22, disinfectant was used to sterilize the surfaces.

Figure 2
A. baumannii on the surfaces of different instruments were removed after sterilization. The heat map shows the colony counts (cfu/cm2) on the surfaces of 10 objects before (Pre, between 8 am and 9 am) and after (Post, between 11 am and 12 am) the surfaces were cleaned using the detergent on day 3 to day 5 and sterilized using the disinfectant on day 6 to day 8.