Microbial Ecology and Antibiotic Susceptibility Profile of Germs Isolated from Hospital Surfaces and Medical Devices in a Reference Hospital in Douala (Cameroon)

Background: The hospital environment is largely contaminated with pathogenic microorganisms. This colonization is a threat for hospitalized patients, especially in high-risk services. The purpose of this study was to identify the germs found on surfaces and medical devices in some departments of the General Hospital of Douala, and to establish their susceptibility profile to most commonly used antibiotics in this health facility. Results: We collected 114 surface and medical device samples, and seeded different culture media for Gram-positive and Gram-negative aerobic bacteria. Of the total samples, 108 were positive and 137 bacterial strains were isolated. The colony count revealed a high rate of contamination. Enterobacter cloacae was the most represented specie (53.3%), followed by Pseudomonas aeruginosa (22.6%) and Klebsiella pneumoniae (6.6%). Various coagulase-negative Staphylococci have been isolated in some departments, as well as Cryptococcus laurentii and molds. The isolated strains showed low susceptibility to the antibiotics tested. Enterobacter cloacae showed low susceptibility for all tested molecules, except for carbapenems with rates ranging from 82% to over 94% in Maternity, Intensive Care and Neonatology units. The strains coming vironment should be regular in critical areas in order to reinforce measures to prevent diffusion of multi-resistant bacteria.


Introduction
Microorganisms largely contaminate the hospital environment; this contamination is variable quantitatively and qualitatively from one institution to another and in a same establishment according to the services. It constitutes a risk factor for the occurrence of nosocomial infections, a real threat, both for the already precarious health of patients and that of the nursing staff and visitors. These infections increase health care costs and length of stay; and are the major cause of mortality and morbidity in hospitalized patients [1] [2]. Although it is difficult to establish a direct link between environmental contamination and the occurrence of nosocomial infections, several studies have shown that microorganisms of human and/or environmental origin contaminate hospital surfaces, and play an important role in the occurrence of these infections [3] [4]. The microorganisms involved in the hospital environment are most often multiresistant to antibiotics, and the main source of diffusion of highly pathogenic strains in services [5] [6] [7] [8]. Microbiological monitoring of the environment in health facilities is part of preventing the transmission of nosocomial infections. The microbiological controls of the environment are one of the measuring tools that make it possible to evaluate a starting situation and the effectiveness of corrective measures, they must be implemented in a relevant way and obey very precise objectives while avoiding the inflation of useless analyses, consuming time and financial means [9] [10] [11]. Efforts are being made to reduce contamination of the hospital environment by developing air and water monitoring methods, surfaces, food, medical devices, care equipment; and the strengthening of hospital hygiene measures [12]. The objective of this study was to quantify and qualify germs present on hospital surfaces and medical devices, and to study their antibiotic susceptibility profile.

Location and Type of Study
We conducted a descriptive cross-sectional study at General Hospital of Douala (GHD) from 1 st January to 30 th June 2015. This is a tertiary health facility located in the Littoral region of Cameroon. This hospital has a capacity of 320 beds and harbours all the major medical and surgical specialities. The choice of high-risk services and surfaces to be taken was made by convenience and focused on the

Sampling
The most exposed surfaces, and low-traffic areas that often escape daily cleaning have been selected, as well as some medical devices. We excluded ceilings and walls in the samples. These samples were taken between 11 AM and 01 PM using a sterile swab previously moistened with sterile saline 0.9%. The swabs were passed in parallel streaks by slightly turning them, on defined areas of 25 cm 2 .

Seeding
The swabs from the different sampling sites were eluted in the test tubes containing 2 ml of sterile physiological saline to suspend the collected elements. The tubes were then vortexed for one minute, and 40 microliters of the resulting suspension were inoculated by the rake method into the following culture media: Plate Count Agar (PCA), Eosin Methylene Blue (EMB) Agar, Mannitol-Salt Agar, The dishes were then incubated in a bacteriological oven at 37˚C for 24 hours for the bacteria, and 48 to 72 hours at 25˚C for the fungi (SC Agar).

Identification of Colonies
Observation of petri dishes after incubation allowed for colony counting and macroscopic identification (shape, size colour). The biochemical and enzymatic identification of the microorganisms was made by seeding a suspension of microorganisms on cards; Vitek2 GN TM for Gram-negative fermenting and non-fermenting bacilli, Vitek2 GP TM for Gram-positive cocci and non-spore-forming bacilli, Vi-tek2 YST TM for yeast and yeast like organisms; followed by incubation and colorimetric reading on the VITEK2 Compact TM 15, an automated microbiology system (bioMerieux SA, France).

Data Processing
The results included the identification number, the date, the time and place of isolation, the identified germs, as well as the antibiotics tested and their susceptibility profile.
The descriptive analysis of the data was made using Epi Info TM version 7.0 and Microsoft Excel 2010 software.

Results
A total of 114 samples were collected from six at-risk and high-risk services, namely: Neonatology in the incubator room, the isolation room and the cloa-  Table 1).
Of these 114 sampling sites, the majority were contaminated and only six (5.26%) were uncontaminated, including the bench in the delivery room, 2 door handles and a neonatal incubator, a bubbler and equipment support in intensive care unit. In all other sectors, no uncontaminated area was noted.
The colony count gave a bacterial density ≥ 10 6 CFU/25 cm 2 , except for the six samples with a sterile culture.
From the 108 contaminated specimens, we isolated 137 microorganisms, with a clear predominance of Enterobacter cloacae ssp cloacae (53.3%), followed by Pseudomonas aeruginosa (22.6%), Negative-coagulase Staphylococcus (8.7%), and Klebsiella pneumoniae (6.6%). For yeasts, 5 strains of Cryptococcus laurentii were isolated, and molds were present in two samples (  departments, the Burn Unit and the Intensive Care Unit. As for Klebsiella pneumoniae, the strains were found in maternity, in neonatology, and in the Burn Unit ( Table 2).
The study of the biochemical and enzymatic characteristics of the most observed strains showed a homology between the strains of Enterobacter cloacae   and those isolated in neonatology showed the same biochemical characteristics for Enterobacter cloacae and K. pneumonia (Table 4).
Regarding antimicrobial susceptibility, isolated strains showed low susceptibility levels. Enterobacter cloacae showed low susceptibility for all tested molecules, except for carbapenems with rates ranging from 82% to over 94% for isolated strains in Maternity, Intensive Care Unit and Neonatology. For the other services, the rates were lower than 37% for the same molecules ( Figure 1). The strains coming from the Haematology Protected Ward were practically resistant to all antibiotics, the highest susceptibility rate being towards quinolones (50% for ofloxacin) (Figure 1). The susceptibility of K. pneumoniae to the antibiotics tested is variable, ranging from 0% for amoxicillin-clavulanic acid in Maternity, to 100% for Carbapenems and Amikacin in Maternity and Neonatology ( Figure   2   for Carbapenems and Fluoroquinolones at 85% and 90%, respectively ( Figure   3). For Negative-Coagulase Staphylococcus, all the strains isolated in the Operating Room and in Neonatology were resistant to Oxacillin, and also S. carnosus and S. sciuri isolated in Maternity and Burn Unit, respectively. Concerning the susceptibility of C. laurentii to antifungal agents, the strains tested were sensitive to Fluconazole, Econazole and Ketoconazole, and resistant to Amphotericin B and Nystatin.

Discussion
The hospital environment is a real reservoir of microorganisms involved in     our study, this may be due to the irregularity of the disinfection of these instruments. The swabbing method used may also play a role in the results of bacterial densities [11]. The high rate of contamination of these critical areas is paradoxical because they are protected areas, normally with low circulation of staff and visitors.
E. cloacae ssp cloacae is the most recovered species, although enterobacteria have little resistance to desiccation, hence their low presence in the environment [14].  Neonatal Intensive Care Unit in Switzerland [15]. This transmission of E cloacae through patients transferred from one hospital to another has also been proven in some France hospitals that have received patients transferred from Morocco [16]. Some strains of K. pneumoniae have been isolated in Neonatalogy, Maternity and Burn Unit. Klebsiella pneumoniae may persist in the environment alone or in combination with Pseudomonas aeruginosa in mixed biofilms [17]. The P. aeruginosa were found in the Burn and Intensive Care Units, they are microorganisms for which humidity and temperature play an important role for their survival in the environment. Some authors have shown that Pseudomonas aeruginosa can survive from few months to several years on hospital surfaces [18].
Because of this ability to grow in wetlands, it is frequently found in the services of burn patients and Intensive Care, with varying susceptibility to antibiotics [19]. Isolated Pseudomonas in health facilities typically produce biofilms, and some genotypes are more productive than others. These biofilms are the starting point of the diffusion in services in taps and bedside tables; and can infect fragile patients such as burn victims [20]. Kominas et al. has demonstrated the transmission of germs, including Pseudomonas between health care staff and patients, and patients to patients in Intensive Care and Burn units [21].
For cocci, only coagulase-negative Staphylococci have been isolated, unlike other studies where S. aureus is predominant [14]. The presence of these SCN has been demonstrated on keyboards in permanent contact with the staff's bare fingers, while the use of gloves for other activities is systematic in these services [3]. Cryptococcus laurentii is the only yeast isolated in different services. This exclusivity may be due to the use of the Vitek 2 TM automatic system for the identification of yeasts, because of the cross reactions between its capsular antigens and those of C. neoformans, identification errors are often observed as demonstrated by Xiao et al. [22]. origin, however they can survive a few days in the hospital environment and be transmitted by the hands of caregivers and surfaces [14].
Although this is only a phenotypic identification, the biochemical homology between strains isolated in some services suggests transmission by staff and patients in the mother and child sector, and between the Operating Room and the Intensive Care Unit.
The susceptibility pattern shows that isolated strains have a low susceptibility rate for antibiotics commonly prescribed in services. Apart from Carbapenemes, which have good activity, resistance levels were high with respect to other molecules, particularly in the protected hematology ward. Chapuis et al. showed the role of the environment in the diffusion of broad-spectrum betalactamase-producing E. cloacae strains in a hematology service, both in the protected area and in the unprotected area [23].

Limitations
Limited ressources, as well as local technical platform did not allow us to perform molecular tests in order to type strains circulating in the hospital. In addition, since the GHD laboratory is specialized in clinical biology, we have not yet acquired new techniques for sampling and analyzing samples taken from the environment.

Conclusion
The investigated surfaces and medical devices of the Douala General Hospital were found to be highly contaminated by pathogenic environmental bacteria and certain yeasts; this allows us confirming the existence of a microbial ecology probably implicated in the occurrence of nosocomial infections. Isolated bacteria are weakly susceptible to the most commonly used antibiotics in these at-risk services. We recommend the reinforcement of staff, spaces and reusable medical devices hygiene, throughout the hospital and particularly in the high-risk areas.
Microbiological controls of the environment should be regular in critical areas in order to reinforce measures to prevent diffusion of multi-resistant bacteria.

Author's Contribution
COE and DA drafted the manuscript; COE and HNL coordinated the study; CMN, JPNM and JB collected data and participated in its design. All the authors read and approved the final manuscript.

Funding
No funding was received in relation to this study.

Ethics Approval
This study was conducted in accordance with ethics directives related to research