Antimicrobial susceptibility patterns of Staphylococcus aureus and coagulase negative staphylococci isolated from humans in Nairobi , Kenya

1 Department of Pharmaceutical Microbiology, Muhimbili University of Health and Allied Sciences (MUHAS), P. O. Box 65013, Dar es Salaam, Tanzania. 2 Kenya Medical Research Institute (KEMRI), Center for Virus Research (CVR), P. O. Box 54628-00200, Nairobi, Kenya. 3 Department of Botany, Jomo Kenyatta University of Agriculture and Technology (JKUAT), P. O. Box 62000, Nairobi, Kenya. 4 Kenya Medical Research Institute (KEMRI), Centre for Microbiology Research, P. O. Box 19464, Nairobi 00202, Kenya.


INTRODUCTION
Staphylococcus aureus is a common cause of both community and hospital-acquired infections.Clinical syndromes associated with severe disease include bacteraemia, pneumonia, endocarditis, septic arthritis, osteomyelitis and deep abscess formation (Enright et al., 2000;Loir et al., 2003, Abdulgader et al., 2015).Mortality from invasive S. aureus disease was high during the pre-antibiotic era.However, the introduction of penicillin in the 1930s had a dramatic impact on the treatment of S. aureus infections.The semisynthetic penicillin methicillin was introduced in 1959 to overcome the problems that arose from the increasing prevalence of penicillinase-producing isolates of S. aureus resistant to penicillin G and penicillin V.However, methicillinresistant S. aureus (MRSA) strains rapidly emerged and became a major clinical problem within hospitals during the 1960s in Europe and the 1970s in the United States and elsewhere (Enright et al., 2000).MRSA was first reported in England in 1961, shortly after its introduction (Mulvey et al., 2001).Worldwide reports show that MRSA strains are resistant to most other classes of antimicrobial agents and are susceptible only to glycopeptides and a few new investigational drugs (Enright et al., 2000;Lee, 2003).
MRSA is common also in the African region (Falagas et al., 2013).From an African multicentre study by Kesah et al. (2003), methicillin resistance was detected in 213 (15%) of the 1440 isolates tested.In another study, the rate of MRSA was relatively high in Nigeria, Kenya and Cameroon (21 to 30%) and below 10% in Tunisia, Malta and Algeria.More than 60% of MRSA were multidrug resistant, with relatively high resistance to erythromycin, gentamicin and oxacillin.Fusidic acid, co-trimoxazole, rifampicin and ciprofloxacin exhibited moderate efficacy.All MRSA isolates were sensitive to vancomycin (Kesah et al., 2003).In a 1995 study conducted in Nairobi, it was found that about 90% of patients admitted in burn units were infected with MRSA and thereby significantly increasing duration of stay in hospital and treatment cost (Muthotho et al., 1995).
The determination of antimicrobial susceptibility of a clinical isolate is often crucial for the optimal antimicrobial therapy of infected patients.This requirement is crucial especially at a time when reports indicate that resistance is increasing and there is emergence of multidrug resistant microorganisms.Standard procedures and breakpoints have been defined to predict therapeutic outcome both in time and at different geographic locations (Fluit et al., 2001).
Monitoring of multidrug resistant methicillin resistant S. aureus is therefore an important public health aspect in Kenya, where MRSA may increase the duration of stay in hospital and treatment costs.In addition, drugs used to treat MRSA are not readily available and affordable in this country.
The aim of this study was to determine the proportion of MRSA and investigate the antimicrobial susceptibility patterns of both coagulase positive and coagulase negative staphylococcal strains isolated from humans in Nairobi, to commonly used antibiotics and vancomycin.

MATERIALS AND METHODS
This was a descriptive and cross sectional study of staphylococci isolates obtained from studies on wounds and blood borne infections and stored at the Centre for Microbiology Research-Kenya Medical Research Institute.

Identification of Staphylococcus species
Samples were subcultured aerobically and growing colonies were Gram stained.Gram-positive cocci clusters were tested for production of catalase and coagulase enzymes and confirmed by API Staph Identification system (BioMerieux Inc., Durham, USA).Confirmed isolates were stored in Tryptic soy broth containing 15% glycerol and frozen at -80°C until further processing.

β-Haemolysis test
Isolates confirmed as S. aureus were inoculated onto blood agar and incubated for 18 to 24 h.Isolated colonies were picked and resuspended in sterile normal saline (0.85%) to give a final inoculums equivalent 0.5 McFarland turbidity standard before inoculating them on Mueller Hinton agar (Oxoid, Hampshire, UK) plates each containing 20 ml of the medium to attain a uniform depth of 4 mm.
The antibiotic disks were placed on each agar plate at equal distance from each other and plates were incubated aerobically at 35°C for 16 -18 h, except for vancomycin, methicillin and oxacillin which were incubated for 24 h.Staphylococcus aureus ATCC 25923 and E. coli ATCC 35218 were used to control for growth of bacteria and efficacy of antibiotic disks.The size of zones of inhibition were recorded and interpreted according to CLSI standards (CLSI, 2015).

Identification of Staphylococcus species
A total of 133 staphylococcal isolates were studied.Out *Corresponding author.E-mail: sangeda@gmail.com.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License of these, 110 (82.7%) were S. aureus (SA) and 23 (17.3%) coagulase-negative staphylococci (CoNS).

β-Haemolysis in Staphylococcus species studied
In total, 88% of all isolates were β haemolytic and the rest were non hemolytic.S aureus isolates were significantly more β hemolytic in comparison with the CoNS isolates (Table 1).

Antimicrobial susceptibility of Staphylococcus species
All isolates were highly sensitive to vancomycin (100%) but highly resistant to penicillin G (91.7%).The order of increasing resistance was penicillin G > augmentin > tetracycline > erythromycin = SXT > oxacillin > methicillin > ciprofloxacin > chloramphenicol > gentamicin > clindamycin > vancomycin (Table 2).Isolates were relatively more sensitive to gentamicin and clindamycin showing resistance in only less than 30% of all isolates.
There was significantly higher resistance to oxacillin in coagulase-negative staphylococci than in S. aureus.The percentage resistance of S. aureus and coagulasenegative staphylococci to oxacillin was 29.2 and 55.7% (P value = 0.0003), respectively.Overall resistance to penicillin and augmentin was significantly higher for S. aureus than coagulase-negative staphylococci (P values = 0.0001 for both).MRSA were detected by determining the susceptibility to oxacillin or methicillin.As shown in Table 3, SA isolates were more resistant to oxacillin than methicillin (33.6 and 31.8%respectively).In contrast, CoNS showed a different trend with oxacillin and methicillin isolates at the proportion of 82.6 and 47.8%, respectively.Out of 12 drugs that were tested, only vancomycin completely inhibited the growth of all isolates tested.Only 7% of all isolates were completely sensitive to all antibiotics (Tables 4 and 5).Sixty three percent of all isolates were multidrug resistant (those resistant to three or more drugs).The remaining 30% of the isolates were resistant to one or two drugs (Table 5).Approximately 8% of all isolates were resistant to all the tested drugs except vancomycin.Other isolates were resistant to one or two drugs (9.7 and 20.3%, respectively).Isolates resistant to between three and ten drugs varied between 4.5 and 12% of all isolates (Table 5).Considered individually, the proportion of multidrug resistant (MDR) isolates in S. aureus and coagulase negative was 58 and 87%, respectively.

DISCUSSION
This study shows that the commonly available drugs such as penicillin, augmentin, tetracycline, erythromycin and SXT are no longer reliable in treating nosocomial and community acquired staphylococcal infections.For instance, out of all isolates tested, 11 were resistant to all antibiotics except vancomycin.Some reports indicate increase of vancomycin intermediate/resistant S. aureus (VISA/VRSA) isolates worldwide (Lowy, 2003;Monaco et al., 2016).Vancomycin resistance became of more concern since the demonstration of successful transfer of the vanA gene from enterococci to S. aureus under laboratory conditions (Noble et al., 1992) and after reports of staphylococcal isolates resistant to vancomycin in some countries (Hiramatsu et al., 1997a, b;Sieradzki et al., 1999;Chang et al., 2003;Whitener et al., 2004;Palazzo et al., 2005;Lee et al., 2015).Because isolates in this study did not show resistance to vancomycin, it is advisable to limit the use of glycopeptides to treat only MRSA infections so as to reduce the selective pressure and likely emergence of resistance to this class of antibiotics.Decreased staphylococcal susceptibility to vancomycin is not due to transfer of vanR genes from vancomycin-resistant enterococci (VRE) or to small colony variants, as noted in staphylococci for other antimicrobial agents (Mitsuyama et al., 1997) but appears to be a gradual selection process due to treatment pressure.Glycopeptideresistant mutants of S. aureus have been experimentally selected by increasing the levels of vancomycin present during in vitro growth (Daum et al., 1992;Sieradzki and Tomasz, 1997).The use of avoparcin, another member of the glycopeptide class of antibiotics, as a growthpromoting agent in the production of food animals is often cited as playing a role in the spread of glycopeptideresistant microorganisms (Aarestrup et al., 1996;Van den Bogaard et al., 1997;Witte, 1997, Economou andGousia, 2015).Although, it is well recognized that vancomycin resistance is more prevalent in the United States than in Europe, it has not been explained why avoparcin usage fails to correlate with the different epidemiologies of resistance between the two continents; avoparcin was never approved for use in animals in the United States, in contrast to its broad use as a growthpromoting agent in Europe (Donnelly et al., 1996;Leclercq and Courvalin, 1997).There is no explanation yet, to this continental variation and more theories are required to explain the difference in the glycopeptide resistance.Based on the outcome of this study and available literature, the use of avoparcin should also be controlled to avoid spread of resistance.In order to avoid the emergence of glycopeptide resistance in currently susceptible staphylococci isolates, the use of avoparcin in Kenya should also be highly controlled.The fact that the use of glycopeptides should be controlled is supported by a number of reports.According to a study by Tenover et al. (1998) glycopeptide-intermediate S. aureus (GISA) isolates represented mutants selected in vivo with increased resistance to glycopeptides as a result of prolonged exposure of the organisms to constant levels of vancomycin in an opportune environment.
In the current study, all isolates had vancomycin zones greater than 14 mm.According to CLSI guidelines, any staphylococcal isolates with zones diameters of 14 or less should be tested by an MIC method.When it is determined that isolates have elevated MIC to vancomycin (>= 4µg/ml), these should be sent to a reference laboratory (CLSI, 2015).According to a study by Tenover and et al. (1998), the disk diffusion testing, which is widely used around the world does not differentiate strains with reduced susceptibility to vancomycin from susceptible strains (MIC range, 0.5 to 2 µg/ml).Accordingly, the disk diffusion test should not be used alone for testing staphylococci resistance with vancomycin.The same study suggests that disk diffusion tests with another glycopeptide, teicoplanin may be of value for identifying isolates with reduced susceptibility to glycopeptides.They also recommend the use of vancomycin agar screening test as method for testing staphylococci MICs and detect isolates with reduced glycopeptide susceptibility (Tenover et al., 1998).
The isolates showed a decreasing trend of resistance in the order, penicillin G > augmentin > tetracycline > erythromycin > SXT > oxacillin > methicillin > ciprofloxacin > chloramphenicol> gentamicin > clindamycin > vancomycin.Similar trends have been observed in South Africa (Marais et al., 2009).This indicates that the antibiotic class of penicillins has higher chances of treatment failure and should not be recommended for treatment of staphylococcal infections.Ciprofloxacin, chloramphenicol, gentamicin and clindamycin should be sought as alternative treatment for MRSA infections.Vancomycin is the most effective drug of all the tested antibiotics when β lactams are ineffective (Shakibaie et al., 2002;Casey et al., 2007;Gad et al., 2010).However, due to poor tissue diffusion and moderate bactericidal activity, vancomycin can be combined with rifampicin for deep-seated infections (Aubry-Damon et al., 1998).
According to the overall proportions of resistant isolates in this study, the authors arbitrary classified the antibiotics with respect to these isolates as highly resistant if percentage resistance is between 50 and 100% as was the case with penicillin and augmentin.Tetracycline, erythromycin, SXT, oxacillin, methicillin, ciprofloxacin, chloramphenicol and gentamicin were moderately resistant (resistance between 25 and 49%).Clindamycin was the only drug to which less than 25% of the isolates were resistant.Interestingly, all isolates were 100% susceptible to the antibiotic vancomycin.
A notable phenomenon with coagulase-negative staphylococci is that they are significantly more resistant to most antibiotics and particularly to oxacillin as compared to S. aureus.This is consistent with previous studies by Reynolds et al. (2004) where 76% of coagulase-negative staphylococci isolates were oxacillin resistant when compared with 42% of S. aureus.In another study, rates of oxacillin resistance among S. aureus and CoNS isolates were 59.3 and 78.5%, respectively (Johnson et al., 2003).The mec gene responsible for S. aureus resistance is postulated to have originated from a different species of staphylococci.Although, many methicillin-resistant strains appear to be descendants of a limited number of clones, some appear to be multi-clonal in origin, suggesting the horizontal transfer of mec DNA (Archer and Niemeyer, 1994;Abdulgader et al., 2015).Hence it is possible that the higher resistance to oxacillin in these coagulase negative isolates could be a source to transfer mec gene to other staphylococci isolates.This was previously confirmed in studies by Wielders et al. (2002).
MRSA isolates often are multiply resistant to commonly used antimicrobial agents, including erythromycin, clindamycin and tetracycline, a situation that was evident in this study.MRSA isolates in this study were not only resistant to β-lactam antibiotics but also to chloramphenicol, clindamycin, ciprofloxacin, erythromycin, gentamicin and sulphamethoxazoletrimethoprim.
Both S. aureus and coagulase-negative staphylococci isolates from humans were highly resistant to penicillin and augmentin.However, about 9% of these isolates were sensitive penicillin.Treatment with these drugs should only be prescribed after sensitivity testing to penicillin.
Indiscriminate use and sale of antimicrobials, sale of antibiotics without prescription, sale of under dose preparations, brand substitution and self-medication can enhance the development of drug resistance (Indalo, 1997;Shakibaie et al., 2002, Roess et al., 2013).Therefore, to control the spread of MRSA, it is advised that both antibiotic use regulation and contact preventions be strictly observed.Similarly, culture surveillance is an important control measure (Farr, 2004;Drees et al., 2016).
Future work is required to determine the clonal relationship among MRSA isolates from a wider area of study, which can be used as epidemiological reference tool during MRSA outbreaks in hospitals and also aid in management and control of these infections.Using molecular tools, the relatedness between S. aureus and coagulase negative staphylococci can possibly indicate the ease with which staphylococci can transfer resistance genes among different genus.

Conclusion
Although, the study isolates were multidrug resistant, there was no isolate which was vancomycin resistant.The fact that 11 (8.3%) of all isolates were resistant to 11 out of 12 tested antibiotics is an alarming situation.Vancomycin should therefore be controlled in hospitals to avoid quick emergence of resistance to this life saving drug.There is a relatively high proportion of oxacillin resistance isolates among coagulase negative staphylococci, these may be responsible for the transfer of MecA gene responsible for oxacillin resistance to the more pathogenic S. aureus strains.Therefore, medical institutions should regularly perform vancomycin agar screening to determine any emergence of glycopeptide and other antibiotics resistance among staphylococci isolates.

Table 3 .
Comparison of overall resistance of S. aureus and coagulase-negative staphylococci.

Table 4 .
Proportion of isolates resistant to one or more of the eleven antibiotics tested.

Table 5 .
Summary of multidrug resistance for all S. aureus and CoNS isolates.