Phenotipic Prevalence of Antibiotic Resistance in Gabon

Background: The increasing phenomenon of bacterial resistance to antibiotics is a real public health problem. The main causes are poor management of hygiene and water quality, but also the use of antibiotics without precaution. The objective of this study was to isolate and determine the antibiotic resistance profile of the different bacteria found in the main hospitals and bacteriology laboratories in Gabon. Methods: 6034 samples were taken from hospitals in seven main cities of Gabon, and analyzed according to the usual techniques. The pathogenic strains were identified by Matrix-Assisted Laser Desorption Ionization-Time Of Flight Mass Spectrometry. Antimicrobial susceptibility testing was performed by the agar disc diffusion method, according to the Antibiogram Committee of the French Society for Microbiology guidelines. Results: 974 pathogenic bacterial strains were found, including 890/974 (91.4%) Gram-negative bacilli. The systematic antimicrobial susceptibility testings identified 160/974 (16.4%) multi-resistant strains. Escherichia coli was the most represented species. 12.5% - 25% of Escherichia coli, Klebsiella pneu-How moniae, Enterobacter cloacae, and Citrobacter sedlakii strains were resistant to amoxicillin + clavulanic acid, third and fourth generation cephalosporins. Aminoglycoside resistance rates of 8.5% - 19% were also noted. 4.5% to 25% of the bacteria found were resistant to quinolones and cotrimoxazole. Resistance rates to carbapenems ranged from 1% to 10.5%. 16% of Staphylococcus aureus were methicillin-resistant (MRSA). Rates of extended spectrum be-ta-lactamase-producing enterobacteriaceae (ESBL-PE) ranged from 2.5% to 25%. Conclusion: This study showed an increasing evolution of bacterial resistance to antibiotics that are spreading throughout Gabon. This constitutes a threat to the health of Gabonese population.


Introduction
Antibiotics have significantly reduced mortality due to bacterial infectious diseases during the 20th century. However, the remarkable efficacy of these molecules has motivated their massive and repeated use in human and animal health.
This created selection pressure on bacteria, resulting in the emergence of resistance [1].
Beta-lactams are the molecules of choice for the treatment of bacterial infections. But their systematic and abusive use has led to resistance to these antibiotics. More and more bacteria are becoming resistant to broad-spectrum beta-lactams, such as third and fourth generation cephalosporins (C3G and C4G) and carbapenems [2]. The inappropriate use of antibiotics for prophylaxis varies between 40% to 75%, depending on the various studies, and 30% to 75% of patients with pulmonary infiltrates receive antibiotics for a non-infectious cause [3] [4] [5]. The link between the use of antibiotics and the emergence of bacterial resistance is clear: the indiscriminate use of antibiotics promotes the emergence of bacterial resistance [1].
Hospital practitioners must increasingly face the problem of the treatment of infections with methicillin-resistant Staphylococcus aureus (MRSA) and of reduced sensitivity to glycopeptides (GISA), to glycopeptides-resistant enterococci (GRE), to multi-resistant Gram negative bacilli [6].
Some bacteria have become multi-resistant, highly resistant, even totally resistant to antibiotics [7]. This phenomenon, which is constantly increasing, often places physicians in a therapeutic impasse, having as consequences an increase in the duration of hospital stay and a higher cost of hospitalization. Antimicrobial resistance has become a public health problem worldwide. is essential to guide optimal empirical treatment in critically ill patients. Antimicrobial susceptibility patterns and bacterial spectra vary geographically, highlighting the importance of local surveillance data [8]. In industrialized countries, these data are available at regional, national and international levels, as shown by the database of the European surveillance system for antimicrobial resistance [9]. In intermediate countries, such as Latin America [10] and Asia [11] [12], good quality data also exist. However, epidemiological data are scarce in Africa, particularly in Central Africa [13] [14].

Knowledge of the epidemiology of infection and antibiotic resistance patterns
In Gabon, there are poor data on the epidemiology of resistance at a national level, despite some studies on antimicrobial resistance [2] [8] [15] [16].
Based on these observations and background, we propose for the first time to conduct a prospective study in the main cities of Gabon, outlining a national map of the prevalence of bacterial resistance to antibiotics.

Settings
This is a multicenter prospective study carried out over a period of 27 months

Inclusion and Exclusion Criteria
The aim of our study was to carry out analysis on all types of biological samples from patients of both sexes, hospitalized or ambulatory, and anal swabs from inpatients were intended for research of the carriage of multi-resistant bacteria (MRB). Samples for research of mycobacteria, chlamydia or mycoplasma were excluded from this study.

Samples
The samples were all types intended for bacteriological analysis, and meeting the inclusion criteria defined above. They were from inpatients or outpatients of all ages and both sexes. Only the urine samples were stored at 4˚C for possible deferred analysis if necessary. All other samples were analyzed upon receipt at the  laboratory. Samples were collected according to protocols commonly used, specific to each type distributed to all hospital departments. They were sent to the laboratory according to these protocols. During the same period, in order to research the carriage of multi-resistant bacteria, anal swabs were collected from hospitalized patients for more than 48 hours without digestive pathology.

Isolation of Bacterial Strains
The samples were seeded on specific solid culture media (chocolate + polyvitex agar, Columbia + sheep blood agar, BCP (Bromo-cresol-purple) agar, EMB (eosin methylene blue) agar, mac Conkey agar, blood + ANC (nalidixic acid-colistin) agar, SS (Salmonella-Shigella) agar, Hektoen agar according to the type of sample and/or the suspected species. The petri dishes were incubated 24 to 48 hours at 37˚C under an atmosphere of 10% CO 2 for the isolation of anaerobic species, and in aerobic conditions for the others.

Identification and Antimicrobial Susceptibility Testing
The suspicious colonies were identified by the Api staph, Api 20E and Api 20NE systems and by the Vitek 2 compact automated system (BioMérieux, Marcyl'Etoile, France).
The antimicrobial susceptibility testings were performed by the Vitek 2 compact.

Search for Extended Spectrum Beta Lactamases (ESBL)
Research for ESBL was based on the highlighting of a synergy between a disc of C3G (cefotaxime, ceftazidime) and/or aztreonam, and a disc of amoxicillin (or ticarcillin) + clavulanic acid. The discs were arranged equidistant from 3 cm center to center.
The appearance of a characteristic image of synergy between the discs, also called "champagne cork", indicated the occurrence of an ESBL (double disc synergy method) [18].
Since the synergy can be masked in cephalosporinase hyper-producing isolates, its detection was facilitated by bringing together the cefepime and/or aztreonam discs with that containing clavulanic acid, or by performing the double-disk synergy test using cloxacillin-supplemented medium (250 mg/L).

Carbapenem Resistance Assessment
If the inhibition zone diameter was less than the susceptibility breakpoint around

Detection of Methicillin-Resistant Staphylococci
The research for methicillin resistance was performed by bringing a disc of cefox-

Glycopeptide Resistance Assessment
The research for resistance to glycopeptides of Staphylococcus aureus strains was carried out exclusively by the Vitek 2 compact automated system on one hand.
On the other hand, the measurement and interpretation of the inhibition zone diameters made it possible to detect the presence or not of Enterococcus strains of the vanA phenotype (resistant to teicoplanin and vancomycin), and/or vanB phenotype (resistant to vancomycin).

Search for Associated Resistances
For all ESBL-producing Enterobacteriaceae (ESBL-PE), carbapenem-resistant bacterial strains IMP-R (imipenem-resistant) and ETP-R (ertapenem-resistant), and Methicillin-Resistant Staphylococcus aureus (MRSA), the resistance to other antibiotics such as quinolones, aminoglycosides and β-lactams have been sought by performing antimicrobial susceptibility testing such as defined above.

Statistics
Statistical analysis was performed using R software.
We compared different parameters using the Chi-square (χ 2 ) method.
Differences were considered statistically significant at 0.05 confidence level.
The resistance rates of Gram-positive cocci are shown in Table 3.

Discussion
Our study revealed a high predominance of urine among all types of samples. This is in line with previous studies which established that urinary tract infections are among the most common bacterial infections contracted in the community and in Table 4. Antibiotic resistance rates (in percentage) of clinical enteriobacteriaceae vs enterobacteriaceae of carriage.   In regard to specific abundance, E. coli (34.67%) for Gram-negative bacilli, and S. aureus (42%) for Gram-positive cocci, were the most representative species.
These proportions corroborate those of Rerambya et al. at the Libreville national public health laboratory [23]. Indeed, the endogenous origin (that is to say coming from the patient himself) of most of the bacteria responsible for the urinary tract infections, places these two species, frequently encountered in the digestive tract, as the most potentially uro-pathogenic [24]. In addition, a study carried out by Guibert et al. established that E. coli is responsible for 80% to 85% of cystitis [25].
According to the EARS-Net (European Antimicrobial Resistance Surveillance Network) data for 2019, our study revealed variations in the occurrence of antimicrobial resistance across the country depending on the bacterial species and antimicrobial group.

Beta-lactam antibiotics are the treatment of choice for infections caused by
Gram-negative bacilli. However, we can increasingly note high resistance rates of these bacteria to beta-lactam antibiotics such as third and fourth generation cephalosporins (C3G and C4G) and carbapenems.
In this study, the resistance rate of E. coli to CAZ (C3G) was higher than that observed by K. Rerambiah et al. at the Libreville national laboratory [23].
In the same line with our results, a study carried out in Senegal on infections caused by enterobacteriaceae also showed that the main ESBL-producing enterobacteriaceae (ESBL-PE) were Enterobacter cloacae, E. coli, Kp [26].
The ESBL rates revealed by our study (2.5% to 25%) were in the same propor- Yala et al. also revealed, using the combined disc method, an ESBL rate of 18% at the Hôpital d'Instruction des Armées Omar BONGO ONDIMBA of Libreville [2]. With respectively 44% and 22.4%, the ESBL-producing Kp rates observed in Latin America and in Pacific Asia were higher than those of our study, which are themselves higher than in Europe (13.3%) and North America (7.5%) [27] [28].
The cephalosporinase hyperproduction rates that we established remained low (2% to 5%). These rates were similar to those observed by De Mouy et al. in a multicenter study in France [29].
The high-level cephalosporinase resistance phenotype results in resistance to all beta-lactams, except carbapenems. There may remain an activity of broad spectrum cephalosporins (cefepime, cefpirome). This is a phenotype found mainly in bacteria naturally having an ampC cephalosporinase which can be overexpressed (Enterobacter cloacae, Citrobacter freundii, Escherichia coli, Morganella morganii, other enterobacteriaceae in the same group) [30].
Resistance to carbapenems suggested the emergence of a production of car- bapenemases by germs, although additional phenotypic and molecular tests remained to be confirmed. This resistance leads to therapeutic impasses in patients, since carbapenems remain the last indication in the event of ESBL [31].
In Gabon, apart from a study by Moussounda et al. who observed the fecal carriage of carbapenemases in a patient hospitalized at HIAOBO [15], very little data exist on this problem. However, Mahamat et al. observed in three main hospitals in N'djamena (Chad), respective prevalence rates of 2.5% and 6.5% in carbapenemase-producing clinical strains and carbapenemase-producing faecal carriage strains [32].
Regarding to the prevalence of MRSA, similar results to those of our study (16%) were observed in Tunisia by Mastouri et al. (15.5% of MRSA at Monastir University Hospital), with the same resistance rate to gentamicin (18%) [33].
However, our study revealed a higher prevalence of MRSA than that observed by Alabi et al. (5.8%) in a retrospective analysis of antimicrobial resistance at the Albert Schweitzer hospital in Lambaréné (Gabon) [8].
Resistance to glycopeptides does not yet seem to be a real concern in hospitals in Gabon, since our results confirmed a global trend, even if studies had reported the existence of MRSA with reduced susceptibility to glycopeptides (GISA) [34] [35].
The analysis of their resistance to antibiotics in this study confirmed the multi-resistant nature of MRSA, usually known for their ability to resist several antibiotics [36].
Unlike K. Rerambiah et al. [23], Streptococcus agalactiae were all susceptible to beta-lactam antibiotics. This should support physicians in the option of penicillins, or even first-generation cephalosporins, as first choice in antibiotic therapy for streptococci infections. Caution should however be exercised in the choice of macrolides and lincosamides for the treatment of streptococcal infections, given the high rate (50%) of resistance to these antibiotics that we observed.
Compared to streptococci, enterococci were less sensitive to penicillins (33.5% of the resistance rate).
Enterococcus faecalis is naturally resistant to lincosamides. So, the high resistance rate to lincomycin (100%) seemed rather to be a good orientation for identification.
The complete absence of glycopeptide-resistant enterococci (GRE) in our study was a piece of good news because, as with MRSA, glycopeptides are the best therapeutic option for infections caused by beta-lactam-resistant enterococci [37]. While, in Europe, as revealed by the annual (2019) epidemiological report on antimicrobial resistance in the European Union (EU) or Economic European Area (EEA), the percentage of vancomycin-resistant isolates of Enterococcus faecium increased from 10.5% in 2015 to 18 Our study showed no significant difference between the antimicrobial resis- tance rates of enterobacteriaceae of carriage and clinical enterobacteriaceae (p > 0.05). However, it has been shown that colonization is essential to the spread of antimicrobial resistance in the population [43]. A link can therefore be established between the patient's colonization by resistant bacteria and the existence of these resistances among bacteria causing infections.
In addition to being resistant to β-lactam antibiotics, MRSA in this study also was resistant to aminoglycosides and quinolones.
In our study, some ESBL-PE were also resistant to aminoglycosides and fluoroquinolones.
Some carbapenem-resistant Enterobacteriaceae, in addition to being resistant to third-generation cephalosporins, were also resistant to aminoglycosides and fluoroquinolones.
This multi-resistance often leads to therapeutic impasse, especially for carbapenemase-producing bacteria.

Conclusion and Perspectives
This study showed the existing variations, in time and space, on antimicrobial resistance. It has allowed us to confirm the continuous increase of antimicrobial resistance, which implies, by prospective studies, updated knowledge of the prevalence of these resistances. Finally, our study would benefit from being supplemented by molecular analysis of resistant strains, in order to identify and determine a national epidemiology mapping of the various resistance genes in Gabon.
So much information is essential for physicians to choose an appropriate antibiotic treatment.