UNIVERSITY OF GHANA  

COLLEGE OF HEALTH SCIENCES  

    

  

  

ANTIMICROBIAL RESISTANCE AND GENETIC DIVERSITY OF  

STAPHYLOCOCCUS AUREUS FROM SURGICAL SITE INFECTIONS AT TWO 

HOSPITALS IN ACCRA  

  

  

  

  

  

BY   

JEANNETTE NANANDAR BENTUM  

(10599788)  

  

  

  

  

  

A THESIS SUBMITTED TO THE DEPARTMENT OF MEDICAL  

MICROBIOLOGY, SCHOOL OF BIOMEDICAL AND ALLIED HEALTH  

SCIENCES, UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF  

THE REQUIREMENT FOR THE AWARD OF MASTER OF PHILOSOPHY  

DEGREE IN MICROBIOLOGY  

  

  

  

  

JULY, 2019   





 

i  

  

  

4/11/2020 



 

ii  

  

DEDICATION   

  

This work is dedicated to my family and to Mr. Kwesi Tipong-Annor Kumi.  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

 

 

 

 

  



 

iii  

  

ACKNOWLEDGEMENT  

First and foremost, I would like to express my sincere gratitude to my supervisors Dr. Beverly 

Egyir and Professor Eric Sampene-Donkor for their support and guidance during the course of 

this project. I would also like to thank Mrs Naiki Attram, Head of the Naval Medical Research  

Unit-3 lab (NAMRU-3) and Dr. Andrew Gordon Letizia, officer in charge of the Naval Medical 

Research Unit Ghana Detachment for the opportunity to conduct this research and their 

guidance through the submission processes. My appreciation also goes to the funders of this 

project, Global Emerging infectious Surveillance (GEIS) and the field staff as well as nurses 

who helped in collecting samples.    

I am indebted to Mrs. Selassie Kumordjie, Miss Clara Yeboah, Mr. George Boateng, Mr. Bright  

Agbodzi, Mr. Eric Behene, Mr. Ronald Essah Bentil, Miss Frema Fiakpui and Miss Jennifer 

Yanney of the NAMRU-3 Lab who helped me with various aspects of my work and without 

whom this work would not have been possible. My gratitude also goes to my colleagues 

especially Mary Abena Attah, Mary-Magdalene Osei and Maame-Esi Dawson Amoah for their 

support.  

Lastly, I would like to thank Mr. Kwesi Tipong-Annor Kumi for his immense support and 

encouragement throughout my course.  

  

  

  

  

 

  



 

iv  

  

 

ABSTRACT  

Background  

Surgical site infections (SSIs) are the most common healthcare-associated infections affecting 

surgical patients. Such infections are often caused by methicillin-susceptible as well as 

methicillin-resistant Staphylococcus aureus. Methicillin-resistant Staphylococcus aureus 

(MRSA) is resistant to the entire class of beta-lactam antimicrobials; which are largely used in 

clinical medicine. Patients infected with MRSAs therefore have limited therapeutic options, 

and this may lead to prolonged periods of hospitalisation and high heath care cost. In Ghana, 

information on SSI as well as the occurrence and prevalence of MRSA and MSSA from such 

infections are scarce. Data on bacteria species recovered from SSI is key for effective 

surveillance and selection of appropriate antimicrobial therapy. This study therefore, 

investigated the proportions of MRSA and MSSA using phenotypic and molecular detection 

tools among patients diagnosed of surgical site infections in two hospitals.   

Aim  

The aim of this study was to determine the antimicrobial resistance patterns and molecular 

characteristics of Staphylococcus aureus detected in patients with surgical site infections at the 

Korle-Bu Teaching Hospital and 37-Military Hospital in Accra.  

Method  

This was a hospital-based cross-sectional study conducted from June to November 2018 at the 

Korle-Bu Teaching Hospital (KBTH) and 37-Military Hospital in Accra. Surgical patients 

diagnosed of SSI were recruited using the Centers for Disease Control (CDC) case definition 

for surgical site infection. Patient demographic data (age, sex, type of operation etc.) and wound 

swabs or aspirates were collected after receiving an informed consent. S. aureus was identified 

using colonial morphology, coagulase testing and the Matrix Assisted Laser Desorption 



 

v  

  

Ionization-Time of Flight Mass Spectrometry (MALDI-TOF-MS). Antimicrobial 

susceptibility testing was performed using the Kirby-Bauer disk diffusion method; measured 

zone sizes were interpreted according the CLSI guidelines. Multiplex PCR was performed to 

detect mecA (methicillin-resistant gene), spa (S. aureus specific gene) and pvl (Panton 

Valentine Leukocidin toxin gene) in the S. aureus isolates. Libraries for illumina sequencing 

were prepared using the Nextera DNA Flex Library preparation kit. Whole genome sequencing 

was done with the MiSeq Illumina sequencer at Noguchi Memorial Institute for Medical 

Research (NMIMR). Genomes were assembled using an in-house pipeline; assembled 

sequences were then uploaded to the centre for genomic epidemiology website 

(http://www.genomicepidemiology.org/) to determine the spa types, sequence types and 

virulence gene content of the S. aureus isolates.  

Results   

A total of 110 patients were recruited into the study, 34 (12.5%) were male and 76 (69.1%) 

were female. Patients between the ages of 25-44 years were highest in the number among the 

patients enrolled. Overall, 13 S. aureus isolates (11.8%; 13/110) were recovered and all were 

resistant to penicillin and susceptible to gentamicin and vancomycin. Cefoxitin resistance 

(4/13; 30.77%) was detected only in isolates from 37-Military Hospital. On the other hand, 

tetracycline (46.15%; 6/13) and norfloxacin (15.38%; 2/13) resistance was recorded at both 

hospitals. Sensitivity of isolates to linezolid (84.62%; 11/13), clindamycin (76.92%; 10/13), 

rifampicin (92.31%; 12/13), co-trimoxazole (92.31%; 12/13) and erythromycin (53.85%; 7/13) 

was very high. The four (30.76; 4/13) isolates resistant to cefoxitin (MRSA) and were also 

positive for mecA by PCR. The predominant S. aureus genotype found in the study was ST152-

t355. The four MRSAs detected belonged to ST152-t355 and ST5-t586 clone. Eight (61.53%; 

8/13) isolates were positive for the Panton Valentine Leukocidin toxin. Twelve other virulence 

genes were detected with haemolysin A and B (hlgA and hlgB) being the most prevalent.  

http://www.genomicepidemiology.org/
http://www.genomicepidemiology.org/


 

vi  

  

Conclusion  

S. aureus isolates recovered were genetically diverse. The detection of ST152 MRSA among 

surgical patients is particularly of interest; this global clone has also been reported in Central 

Europe, the Balkan, Switzerland and Denmark as a community acquired MRSA. Continuous 

surveillance may be required to monitor the spread of these pandemic clones in the hospital 

setting.   

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

  

 

  

  

  



 

vii  

  

TABLE OF CONTENTS   

DECLARATION ........................................................................................................................i  

DEDICATION.......................................................................................................................... ii  

ACKNOWLEDGEMENT ....................................................................................................... iii  

ABSTRACT............................................................................................................................. iv  

LIST OF TABLES .................................................................................................................. xii  

LIST OF FIGURES ............................................................................................................... xiii  

LIST OF ABBREVIATIONS ................................................................................................ xiv  

LIST OF APPENDICES........................................................................................................ xvi  

CHAPTER ONE........................................................................................................................ 1  

1.0 INTRODUCTION ............................................................................................................ 1  

1.1 Background .................................................................................................................. 1  

1.2 Problem Statement ....................................................................................................... 3  

1.3 Justification .................................................................................................................. 4  

1.4 Aim .............................................................................................................................. 5  

1.4.1 Specific Objectives ................................................................................................ 5  

CHAPTER TWO....................................................................................................................... 6  

2.0 LITERATURE REVIEW............................................................................................. 6  

2.1 Surgical site infections.................................................................................................. 6  

2.2 Magnitude of surgical site Infections ........................................................................... 7  

2.3 Economic Burden of Surgical Site Infections.............................................................. 8  

2.4 Microbiology of Surgical Site Infections ..................................................................... 9 



 

viii  

  

2.5 Overview of Staphylococcus aureus.......................................................................... 10 

2.6 Virulence Factors of S. aureus................................................................................... 12 

2.7 Antimicrobial Resistance in Staphylococcus aureus.................................................. 14  

2.7.1 Methicillin Resistance in Staphylococcus aureus ................................................ 14  

2.7.2 Resistance Patterns of S. aureus and MRSA ....................................................... 17  

2.8 Typing Methods for S. aureus .................................................................................... 19  

2.8.1 Pulsed Field Gel Electrophoresis (PFGE)............................................................ 20  

2.8.2 Spa Typing .......................................................................................................... 20  

2.8.3 Multi-Locus Sequence Typing (MLST).............................................................. 21  

2.8.4 SCCmec Typing.................................................................................................. 21 

2.8.5 Whole Genome Sequencing (WGS).................................................................... 21  

2.9 Molecular Epidemiology and Genetic diversity of Staphylococcus aureus ............... 23  

2.9.1 Molecular Epidemiology and Genetic Diversity of Staphylococcus aureus in  

Africa ........................................................................................................................... 25  

CHAPTER THREE................................................................................................................. 27  

3.0 MATERIALS AND METHODS................................................................................... 27  

3.1 Study Design............................................................................................................... 27  

3.2 Study Area .................................................................................................................. 27  

3.2.1 37-Military Hospital............................................................................................. 27  

3.2.2 Korle-Bu Teaching Hospital................................................................................ 28  

3.3 Ethical Clearance ....................................................................................................... 28 

3.4 Recruitment of Study Participants ..............................................................................29 



 

ix  

  

3.4.1 Inclusion Criteria ................................................................................................. 29 

3.4.1.1 CDC Criteria ................................................................................................. 29 

3.4.2 Exclusion Criteria ................................................................................................ 30  

3.5 Sample Collection....................................................................................................... 30  

3.6 Laboratory Procedures............................................................................................... 30  

3.6.1 Sample Processing............................................................................................... 30  

3.6.2 Identification of Bacterial Isolates....................................................................... 30  

3.6.2.1 Gram Stain..................................................................................................... 31  

3.6.2.2 Catalase Testing............................................................................................ 31  

3.6.2.3 Coagulase Testing......................................................................................... 31  

3.6.3 Confirmation of Isolates Identified ...................................................................... 32  

3.6.4 Antimicrobial Susceptibility Testing ................................................................... 33  

3.7 Molecular Techniques ................................................................................................ 34  

3.7.1 DNA Extraction ................................................................................................... 34  

3.7.2 Polymerase Chain Reaction ................................................................................. 34  

3.7.3 Gel Electrophoresis ............................................................................................. 35  

3.7.3.1 Gel Preparation ............................................................................................. 36  

3.7.3.2 Gel loading and Running .............................................................................. 36  

3.7.4 DNA Extraction for Sequencing.......................................................................... 36  

3.7.5 Whole Genome Sequencing................................................................................ 37  

3.8 Statistical Analysis .....................................................................................................38 

CHAPTER FOUR ................................................................................................................... 39 



 

x  

  

4.0 RESULTS....................................................................................................................... 39 

4.1 Demographic Characteristics of Study Participants ................................................... 39  

4.2 Patients Enrolled at the Various Departments/Hospitals............................................ 40  

4.3 Antibiotic Use before and after Operation................................................................. 42  

4.4 Proportion of Staphylococcus aureus Isolated from Study Participants.................... 43  

4.5 Antimicrobial Resistance Pattern of Staphylococcus aureus ..................................... 44  

4.6 Antimicrobial Resistance Profile of MRSA and MSSA............................................ 45  

4.7 Molecular Characteristics of Staphylococcus aureus Isolates .................................... 47  

4.7.1 Molecular Detection of mecA, spa and pvl genes ................................................ 47  

4.7.2 Spa/Sequence type and Toxin Gene Distribution among Staphylococcus aureus  

Isolates ......................................................................................................................... 47  

4.7.3 Virulence Factors for S. aureus isolates.............................................................. 48  

4.7.4 Toxin Genes Present in MRSA and MSSA Isolates............................................ 50  

4.7.5 Antimicrobial Resistance patterns of MRSA and MSSA Isolates....................... 50  

CHAPTER FIVE ..................................................................................................................... 52  

5.0 DISCUSSION ................................................................................................................ 52  

5.1 Proportions of SSI Positive for S. aureus ................................................................... 52  

5.2 Antimicrobial Susceptibility Patterns of S. aureus Isolates ....................................... 53  

5.3 MRSA and MSSA among S. aureus Isolates ............................................................. 56  

5.4 Molecular Epidemiology and Genetic Diversity of S. aureus Isolates....................... 57  

CHAPTER SIX.........................................................................................................................62 

6.0 CONCLUSION, RECOMMENDATION, LIMITATIONS......................................... 62 



 

xi  

  

     6.1 Conclusion .................................................................................................................. 62 

     6.2 Recommendations ...................................................................................................... 62  

   6.3 Limitations .................................................................................................................. 63  

REFERENCES ........................................................................................................................64  

APPENDIX I: Participant Consent Form................................................................................ 76  

APPENDIX II: Parental Consent Form................................................................................... 78  

APPENDIX III: Child Assent Form ........................................................................................ 80  

APPENDIX IV: Case Report Form......................................................................................... 82  

APPENDIX V: NHSN Operative Procedures Followed for 30 and 90 Days.......................... 83  

APPENDIX VI:  DNA Quantification for NGS Using the Qubit Double Strand (ds) High  

Sensitivity (HS) Kit.................................................................................................................. 84  

APPENDIX VII: Tests for Identification and Susceptibility testing of S. aureus ................... 85  

 

 

 

 

 

 

 

 

 

 

 

 



 

xii  

  

LIST OF TABLES  

Table 3.1: Primer Sequences used in the Study....................................................................... 35  

Table 3.2: One millilitre (1ml) Primer mix preparation .......................................................... 35  

Table 4.1: Demographic Characteristics of Study Participants............................................... 39  

Table 4.2: Preoperative Antibiotic Use .................................................................................... 42  

Table 4.3: Post-Operative Antibiotic Use................................................................................ 43  

Table 4.4: Proportions of Patients Positive for Staphylococcus aureus at the two Hospitals. 44  

Table 4.5: Antimicrobial Resistance Profile of Staphylococcus aureus Isolates ..................... 45  

Table 4.6: Antimicrobial susceptibility Patterns of MRSA and MSSA Isolates ..................... 46  

Table 4.7: Distribution of Spa/Sequence type and PVL toxin among Hospitals ..................... 48  

Table 4.8: Toxin Genes Present in MRSA and MSSA Isolates............................................... 50  

Table 4.9: Antibiotype and spa/sequence types of MRSA and MSSA................................... 51  

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

xiii  

  

LIST OF FIGURES  

Figure 4.1: Distribution of Patients Enrolled into the Study ................................................... 41  

Figure 4.2: 2% Agarose gel picture of multiplex PCR for mecA, spa and pvl genes.............. 47 

Figure 4.3: Toxin Genes Present in Staphylococcus aureus Isolates....................................... 49 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

xiv  

  

LIST OF ABBREVIATIONS  

ATTC                American Type Culture Collection  

β-lactam            Beta-lactam  

CA-MRSA        Community-Associated Methicillin Resistant Staphylococcus aureus  

CDC                  Centers for Disease and Control  

CLSI                  Clinical and Laboratory Standards Institute  

CONs                 Coagulase-Negative Staphylococci  

DNA                  Deoxyribonucleic Acid  

ESBL                 Extended Spectrum beta-lactamase  

EUCAST           European Committee on Antimicrobial Susceptibility Testing  

HAI                   Hospital Acquired Infections  

HA-MRSA        Healthcare-Associated Methicillin Resistant Staphylococcus aureus  

KBTH                Korle-Bu Teaching Hospital  

LA-MRSA         Livestock-Associated Methicillin Resistant Staphylococcus aureus  

MDR                  Multi-drug Resistant    

MLST                Multi-locus Sequence Typing  

MRSA                Methicillin Resistant Staphylococcus aureus  

MSSA                 Methicillin Susceptible Staphylococcus aureus  

MSCRAMMS    Microbial Surface Components Recognizing Adhesive Matrix molecules  

NHSN                 National Healthcare Safety Network  



 

xv  

  

NMIMR               Noguchi Memorial Institute for Medical Research  

NNIS                    National Nosocomial Infections Surveillance  

PBP                       Penicillin binding protein  

PBS                       Phosphate Buffered Saline  

PCR                       Polymerase Chain Reaction  

PFGE                     Pulse-field Gel Electrophoresis  

PVL                       Panton-Valentine Leukocidin 

 qPCR                    Quantitative Polymerase Chain Reaction  

S. aureus                Staphylococcus aureus  

SCCMEC               Staphylococcal Cassette Chromosome mec  

spa                          Staphylococcal Protein A  

SSI                          Surgical Site Infections  

ST                           Sequence Type  

TAE                        Tris-Acetate EDTA  

TSST-1                   Toxic Shock Syndrome Toxin -1  

VISA                      Vancomycin Intermediate Staphylococcus aureus  

VRSA                     Vancomycin Resistant Staphylococcus aureus 

WGS                       Whole Genome Sequencing  

γ hemolysin             Gamma Hemolysins  

  



 

xvi  

  

LIST OF APPENDICES  

APPENDIX I: Participant Consent Form................................................................................ 76  

APPENDIX II: Parental Consent Form................................................................................... 78  

APPENDIX III: Child Assent Form ........................................................................................ 80  

APPENDIX IV: Case Report Form......................................................................................... 82  

APPENDIX V: NHSN Operative Procedures Followed for 30 and 90 Days.......................... 83 

APPENDIX VI:  DNA Quantification for NGS Using the Qubit Double Strand (ds) High  

Sensitivity (HS) Kit.................................................................................................................. 84  

APPENDIX VII: Tests for Identification and Susceptibility testing of S. aureus ................... 85  

  

  

  

  

  

  

  

  

  

  

  

  

  

  



 

1  

  

CHAPTER ONE  

1.0 INTRODUCTION  

1.1 Background  

Healthcare-associated infection (HAI) is the most common challenge patients are faced with 

when they are hospitalized (Zimlichman et al., 2013). Across the globe, 14-33% of HAIs have 

been classified as Surgical site infections (SSIs) (Allegranzi et al., 2011; Klevens et al., 2007; 

Labi et al., 2019; Magill et al., 2012; Smyth et al., 2008). The Centers for Disease Control and 

Prevention (CDC) defines an SSI as an infection that occurs at the site of incision within 30 

or 90 days and up to a year (if there is an implant) of a surgical procedure (CDC, 2019). 

Worldwide, approximately 2-5% of surgical patients have been estimated to develop a surgical 

site infection (Smyth et al., 2008). SSIs are worth noting because they pose a threat to most 

patients who undergo surgery by causing delayed wound healing, longer hospitalization, 

increased readmission rates, increased healthcare cost as well as increased morbidity and 

mortality (Awad, 2012; Grimble et al., 2001; Klevens et al., 2007; Merkow et al., 2015; 

Zimlichman et al., 2013). The economic burden alone associated with SSIs have been 

estimated to be over 3 billion US dollars in the US (Zimlichman et al., 2013).  

SSIs are considered preventable infections because most SSIs are caused by microorganisms 

that are a part of the body’s normal flora. However, organisms that are found in the hospital 

environment are also implicated (Awad, 2012; Bastola et al., 2017). Bacteria species known 

to cause SSIs include: Staphylococcus aureus, coagulase negative Staphylococci, 

Enterococcus spp, Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, 

Streptococcus pyogenes, Enterobacter spp, Acinetobacter spp and Proteus spp. Among these 

organisms, Staphylococcus aureus seems to be the most predominant (Dhar et al., 2014; Lilani 

et al., 2005; Mpogoro et al., 2014) causing up to 20% of SSIs (Mangram et al., 1999).   



 

2  

  

S. aureus is a normal flora of the skin, nares, throat and vagina. However, its ecological niche 

is the anterior nares, and at any given time, 20% of the population is persistently colonized 

(Reygaert, 2013). It has inherent virulence factors like the Panton-Valentine Leukocidin 

(PVL) toxin, Protein A, coagulase gamma toxins (HlgA, HlgB and HlgC) and Staphylococcal 

enterotoxins (SEA, etc.) which enable it to colonize and invade the host tissues (Hetem et al., 

2016; Reygaert, 2013). In the early 1940s, penicillin was used to treat staphylococci 

infections, within a few years of use, resistance of S. aureus to penicillin developed. This led 

to the introduction of methicillin which initially solved the problem. However, two years after 

its introduction, resistance to methicillin developed as well (Reygaert, 2013). By 1990, 20-

35% of S. aureus infections in hospitalized patients were caused by Methicillin-Resistant 

Staphylococcus aureus (MRSA). MRSA has become a clinical concern because of its ability 

to be resistant to several antimicrobials (Kahsay et al., 2014). MRSA strains are classified as 

healthcare-associated MRSA (HA-MRSA), community-associated MRSA (CA-MRSA), and 

livestock-associated MRSA (LA-MRSA), depending on whether they are associated with 

hospitals, communities, or livestock (David & Daum, 2010; Gorwitz, 2008; Moellering, 

2012). Community-associated MRSA tend to be susceptible to other classes of antimicrobials 

apart from Beta-lactams (David & Daum, 2010) and carry the PVL toxin believed to be 

associated with increased disease severity and necrosis (Gorwitz, 2008).   

Understanding the spread and evolution of S. aureus is essential to the development of 

effective systems to control its spread (Deurenberg & Stobberingh, 2008). Currently there are 

various typing methods for detecting genetic variation in S. aureus. Thus far, pulsed field gel 

electrophoresis (PFGE) has been the primary method (Deurenberg & Stobberingh, 2008; 

Rodriguez et al., 2015; Sabat et al., 2013). In more recent times, there has been a shift towards 

the use of Staphylococcal Protein A (spa), Multi-locus sequence typing (MLST) and 

staphylococcal cassette chromosome (SCC) mec typing (Deurenberg & Stobberingh, 2008). 



 

3  

  

However, the type of method to use hinges on the problem to be addressed (O’Hara et al., 

2016; Sabat et al., 2013).   

1.2 Problem Statement   

SSIs are the most common healthcare-associated infections and account for up to 33% of all 

hospital-acquired infections in Ghana (Labi et al., 2019). SSIs remain an important clinical 

problem as they cause an estimated number of  8,205 deaths each year, prolong hospital stays 

for patients by at least 11 days, lead to 20% of unplanned readmissions for surgical patients 

and increases healthcare cost (Awad, 2012; Klevens et al., 2007; Merkow et al., 2015; 

Zimlichman et al., 2013). In addition, the healthcare system of countries who have limited 

resources are further burdened (Klevens et al., 2007; Scott, 2009). In the US alone, an 

increased hospital stay has been shown to lead to an average of $2900 to $3000 increase in 

cost for every SSI (Awad, 2012). In Africa, SSI rates of 2.5% to 30.9% have been reported 

(Nejad et al., 2011). Postoperative hospital stay associated with SSI has been found to be 13-

20 days in Tanzania and Ethiopia (Eriksen et al., 2003; Taye, 2005). Other studies conducted 

in Ethiopia rt35have reported a mortality rate of 10.8% for patients with SSI compared to 

3.9% for uninfected patients (Taye, 2005) and a delayed hospital discharge in 14.7% patients 

with SSI (Kotisso & Aseffa, 1998).  

 Saadatian-Elahi, Teyssou, & Vanhems  (2008)  20% of SSIs are caused by S. aureus. Studies 

in Africa have also shown that the predominant bacteria isolated from SSIs is S. aureus 

(Bercion et al., 2007; Eriksen et al., 2003; Fehr et al., 2006; Kotisso & Aseffa, 1998). Amongst 

the S. aureus strains, MRSA is rising in prevalence mainly on the African continent, and is 

implicated in SSIs (Ahmed et al., 2014; Falagas et al., 2013; Iyamba et al., 2014; Kheder et 

al., 2012.). This is of major concern as the multi-drug resistant nature of MRSA makes control 

and management of such infections challenging for physicians. The consequence of this is 

delayed wound healing as well as increased death rates (Cosgrove et al., 2003).   



 

4  

  

In Ghana, it has been reported that MRSA prevalence is low (Donkor et al., 2019), however, 

one of the very few studies conducted across the country found a prevalence as high as of 

54.7% in wound samples (Odonkor et al., 2012). MRSA has also been identified in blood 

samples (Dekker et al., 2016) with two studies showing high prevalence of 86.2% (75/87) and 

86.5%(64/75) (Karikari et al., 2017; Laryea et al., 2014). Additionally, nasal carriage of 

MRSA among 10% of surgical patients has been reported (Egyir et al., 2013) and this is 

worrying as nasal carriage of S. aureus is an established risk factor for SSIs (Levy et al., 2013; 

Mangram et al., 1999). These reports show that MRSA is on the rise and needs to be addressed 

immediately.  

Although S. aureus particularly MRSA has both serious economic and health implications, 

information on its occurrence and spread, resistance patterns and characterization in surgical 

site is limited throughout Africa and especially in Ghana. Data available in Ghana is on S. 

aureus and MRSA from other samples such as blood, urine, pus and wound but little is known 

about surgical sites.  

1.3 Justification   

The outcome of SSI is too high a price to pay and considering the difficulties that arise in 

treating infections associated with MRSA and the consequences, it is important that its spread 

is controlled. To be able to do this, effective control and surveillance programs should be put 

in place to prevent, monitor and regulate the spread of this organism. These measures however 

rely on accurate information on the epidemiology of the organism (Weller, 2000). In addition, 

to make prompt decisions on the appropriate antibiotic to use for effective treatment, 

knowledge about the resistance patterns is essential (Knobler et al., 2003). Early and 

appropriate treatment when put in place will also help to control the spread of the organism 

(Knobler et al., 2003). So far, information on surgical site infections as well as the role of S. 

aureus (MRSA and MSSA) principally, in SSI is scarce. This study will therefore fill an 



 

5  

  

important knowledge gap and help develop potential interventions to reduce the occurrence 

of MRSA.  

1.4 Aim  

The aim of this study is to determine the antimicrobial resistance patterns and molecular 

characteristics of Staphylococcus aureus recovered from patients diagnosed of surgical site 

infections at the Korle-Bu Teaching Hospital and 37-Military Hospital in Accra.  

1.4.1 Specific Objectives  

The specific objectives of the study are to:  

• Determine the proportion of patients diagnosed of SSI positive for S. aureus and the 

antimicrobial resistance patterns of the isolates.   

• Determine the proportion of MRSA and MSSA among the S. aureus isolates.  

• Determine the molecular characteristics of S. aureus isolates recovered.  

  

  

 

 

 

 

 

 

 

  



 

6  

  

CHAPTER TWO  

2.0 LITERATURE REVIEW  

2.1 Surgical site infections   

An infection of a surgical site is a frequent complication of surgery and the commonest 

hospital-acquired infection (Awad, 2012). Although infection control practices comprising 

sterilization techniques, surgical skills and availability of surgical prophylaxis have improved, 

SSIs persists and remain a problem (CDC, 2019). The term SSI was introduced to replace 

surgical wound infections in 1992 (Owens & Stoessel, 2008). Surgical site infections form 

14-33% of the collection of HAIs (Allegranzi et al., 2011; Klevens et al., 2007; Labi et al., 

2019; Magill et al., 2012; Smyth et al., 2008). SSIs are defined as infections that occur at the 

site of an incision within 30 or 90 days or up to a year (if there is an implant) of a surgical 

procedure (CDC, 2019; Mangram et al., 1999). Early efforts to determine the burden of 

hospital-acquired infections led the CDC to establish the National Nosocomial Infections 

Surveillance (NNIS) system which was later replaced by the National Healthcare Safety 

Network (NHSN) to monitor healthcare–associated infections in US acute hospitals (Magill 

et al., 2012; Mangram et al., 1999). For the reason that SSI definitions were not universal, the 

CDCs NNIS system developed a standardized criteria for defining SSI which is now being 

used worldwide (Mangram et al., 1999). This was essential as reporting accurate and 

interpretable SSI rates depended on consistent and standardized definitions. By the CDC 

criteria, SSIs are classified as superficial, deep or organ space in relation to the depth of 

infection (Mangram et al., 1999). Superficial incisional involves the skin and cutaneous tissue 

and occurs within 30 days. Deep incisional involves deep soft tissues and occurs within 60 or 

90 days depending on the surgical procedure performed. Organ space infections occur beyond 

the deep soft tissues and usually have a 90-day surveillance period. However, superficial 

infections account for the majority of SSIs and organ space infections are less commonly 



 

7  

  

encountered. Besides the time period and depth of infection, SSI classification is based on the 

presenting signs and symptoms like purulent discharge at the site of incision, pain, tenderness, 

swelling at the site, redness, heat and fever (CDC, 2019; Mangram et al., 1999). Diagnosis of 

SSI by an attending physician or identification of organisms from the incision by either 

microbiological testing for treatment or clinical diagnosis also form the basis of classification 

(CDC, 2019; Mangram et al., 1999). In more recent times, SSI criterion includes having an 

NHSN operative procedure. This is described as a procedure contained within the NHSN 

operative code mapping, occurs in the course of an operation where the skin or mucous 

membrane is incised at least once, or a reoperation via a previous incision and takes place in 

an operating room (CDC, 2015, 2019).   

SSIs are considered preventable infections. Several measures have been suggested to aid in its 

control and prevention such as appropriate antimicrobial prophylaxis, reducing microbial 

contamination of surgical instruments by sterilization, putting SSI surveillance in place, 

proper skin preparation and decontamination of environmental surfaces (Anderson et al., 

2014; Mangram et al., 1999). Accordingly, most hospitals, especially in developed countries, 

have implemented surveillance systems to detect infections and successfully decrease SSIs 

(Song et al., 2018). The CDC reported a 17% decrease in SSIs between 2008 and 2014 among 

10 different procedures (CDC, 2019). This is probably an indication of how well these systems 

are working.  

2.2 Magnitude of surgical site Infections   

The CDC’s NNIS system reported that between 1986 and 1996, there were 15,523 SSIs 

following 593,344 operations in acute-care hospitals in the US (Mangram et al., 1999). Then 

the NHSN reported a 1.9% SSI rate for operations occurring between 2006 and 2008 (Mu et 

al., 2011). Additionally, a healthcare-associated infection survey conducted by the CDC also 

found that SSIs among in-patient surgeries in the US in 2011 was approximately 157,500 



 

8  

  

(Magill et al., 2014). However, the problem of SSI is universal. SSI incidence rates of 2.2%, 

1.6%, 1.4%, 1.6% and 2.0% have been found in Europe, Germany, England, France and 

Portugal respectively (Fan et al., 2014). In mainland China, it has been reported to be 4.5% 

(Fan et al., 2014). Globally, approximately 2-5% of surgical patients have been estimated to 

develop a surgical site infection (Smyth et al., 2008). There is, however, a substantial 

difference between developed countries and developing countries in terms of this problem. In 

developed countries, SSI is often more controlled probably due to surveillance systems that 

have been put in place, whereas in developing countries, scarce resources, as well as 

inadequate infection control practices may contribute to the problem (Allegranzi et al., 2011; 

Nejad et al., 2011). Data is limited on SSI in Africa, yet, rates of 2.5% to 30.9% have been 

reported in some parts (Nejad et al., 2011). In Ghana, SSIs have been shown to account for up 

to 33% of all hospital acquired infections (Labi et al., 2019) and SSI prevalence as high as 

39% and 40% have been reported (Ameyaw, 2014; Apanga et al., 2014).   

2.3 Economic Burden of Surgical Site Infections   

SSI management imposes a significant burden on patients and the healthcare system of a 

country (Klevens et al., 2007; Scott, 2009). SSIs are also associated with substantial 

morbidity. Patients with SSIs have to be hospitalized for longer times or readmitted to the 

hospital and this increases the healthcare cost for them (Awad, 2012; Grimble et al., 2001; 

Klevens et al., 2007; Merkow et al., 2015; Zimlichman et al., 2013). Apart from that, 

individuals with these infections have a higher risk of dying compared with individuals 

without infections. The economic burden associated with SSIs has been estimated to be over 

3 billion US dollars in the US (Zimlichman et al., 2013), $1,341-$10922 per patient in England 

and an average of $4,544 per day in Europe (Fan et al., 2014). In the US, an increased hospital 

stay has been shown to lead to an average of $2900 to $3000 increase in cost for every SSI 

(Awad, 2012). Additionally, SSIs have been associated with an estimated 8,205 deaths each 



 

9  

  

year, prolonged hospital stays of at least 11 days, 20% of unplanned readmissions and 

increased healthcare cost for patients (Awad, 2012; Klevens et al., 2007; Merkow et al., 2015; 

Zimlichman et al., 2013). In Africa, information about the economic burden of SSI is scarce. 

However, a few studies that have been conducted in some countries like Tanzania and 

Ethiopia, have reported SSI associated postoperative hospital stay of 13-20 days (Eriksen et 

al., 2003; Taye, 2005), a mortality rate of 10.8% (Taye, 2005) and a delayed hospital discharge 

in 14.7% SSI patients (Kotisso & Aseffa, 1998). SSIs and their corresponding economic 

burden have not been studied in Ghana. This may be due to the fact that it has not been 

acknowledged as a major problem. Much attention has been given to diseases such as malaria, 

tuberculosis and HIV. However, the impact of SSIs is great and as such surgical patients who 

develop an SSI need to be given greater attention.  

2.4 Microbiology of Surgical Site Infections  

SSIs are most often caused by the endogenous flora of the patient and the organism isolated 

is dependent on the type of surgery performed (Awad, 2012; Mangram et al., 1999; Owens & 

Stoessel, 2008). However, exogenous sources such as the hospital environment has also been 

implicated (Awad, 2012; Bastola et al., 2017). In the hospital setting, these organisms may be 

acquired by direct contact with hospital staff or other patients and improperly sterilized 

equipment or materials that are used during the surgical operation (Awad, 2012). There are 

several patient-related and procedure-related factors also contributing to the occurrence of 

SSIs. Patient-related factors include age, obesity, underlying diseases such as diabetes mellitus 

(Anderson, 2011; Owens & Stoessel, 2008), and nasal carriage of S. aureus (Levy et al., 2013; 

Mangram et al., 1999). Procedure-related factors include longer duration of surgery 

(Marimuthu et al., 2016; Mpogoro et al., 2014). Organisms causing infection have not 

drastically changed over the last two decades (Akinkunmi et al., 2014; Bhave et al., 2016; 

Lilani et al., 2005; Mangram et al., 1999; Takesue et al., 2017) probably because endogenous 



 

10  

  

and exogenous organisms continue to be the same. Staphylococcus aureus, coagulase-

negative Staphylococci, Enterococcus spp, Escherichia coli, Klebsiella pneumoniae, 

Pseudomonas aeruginosa, Streptococcus pyogenes, Enterobacter spp, Acinetobacter spp and 

Proteus spp are still frequently isolated from SSIs (Mangram et al., 1999). Among these 

organisms, Staphylococcus aureus seems to be the most predominant (Dhar et al., 2014; Lilani 

et al., 2005; Mpogoro et al., 2014) causing up to 20% of SSIs (Mangram et al., 1999) and 

nasal colonization being one of the most important risk factors for the development of an SSI 

(Levy et al., 2013; Mangram et al., 1999; Wertheim et al., 2005). Reports have also shown an 

increased number of SSIs associated with resistant bacteria such as MRSA (Mangram et al., 

1999; Owens & Stoessel, 2008), worsening the burden associated with it.  

2.5 Overview of Staphylococcus aureus   

Staphylococci are a genus of Gram-positive cocci bacteria that belong to the order Bacillales 

and family Staphylococcaceae (Crossley & Archer, 1997). The first identification of 

Staphylococci was by Alexander Ogston from surgical abscesses in the knee joint (Crossley 

et al., 2009; Hetem et al., 2016). Staphylococci are viewed as pairs or grape-like clusters under 

the microscope from which its name was derived (Crossley et al., 2009; Hetem et al., 2016). 

Generally, they grow well on blood agar but some strains prefer a medium with high salt 

concentration such as mannitol salt agar. On blood agar, they appear as 2mm golden coloured 

or white colonies (Crossley et al., 2009; Hetem et al., 2016) and as yellow colonies with yellow 

zones or pink colonies on mannitol salt agar (Maza et al., 2004). Staphylococci are facultative 

anaerobes, non-motile and non-spore forming organisms (Hetem et al., 2016). A characteristic 

feature of Staphylococci is the production of catalase, an enzyme that converts hydrogen 

peroxide to oxygen and water and distinguishes them from closely related members of the 

genus Streptococci which are Gram-positive cocci in chains but catalase-negative (Foster, 

1996). Most often, they are found as normal flora of the skin, nasopharynx, vagina and mucous 



 

11  

  

membrane and as such they are not of concern. Yet, a break in the skin or a compromised 

immune system provides an opportunity to cause disease (Reygaert, 2013). There are over 50 

species that have been described so far (Hetem et al., 2016). These species can be divided into 

two groups based on their ability to clot blood plasma by the production of an enzyme known 

as coagulase (Foster, 1996). Coagulase producing species comprise S. aureus, S. intermedius, 

S. delphini, S. lurae, S. hyicus and S. schleiferi subsp. Coagulans (Sasaki et al., 2010). Species 

that lack this enzyme are known as Coagulase-Negative Staphylococci (CONs) and some 

members that have been known to cause human infections include: S. epidermidis, S. 

saprophyticus, S. haemolyticus, and S. lugdunensis.  S. saprophyticus has been associated with 

urinary tract infections in young female adults (Crossley et al., 2009)  

Staphylococcus aureus, a member of the coagulase-positive group is the most pathogenic in 

this genus (Foster, 1996). During the 1917-1918 Spanish influenza pandemic, it was proposed 

that most of the deaths were a result of bacterial pneumoniae caused by upper respiratory tract 

pathogens: Streptococcus pneumoniae, Streptococcus pyogenes and S. aureus (Humphreys, 

2018; Morens et al., 2008). Similarly, during the First World War, septic wound infections 

that occurred were believed to be due to S. aureus (Hetem et al., 2016). Although S. aureus 

strains are generally coagulase-positive, some may not produce coagulase (atypical) (Foster, 

1996). S. aureus is found as normal flora of the skin and nasopharynx. However, it is mainly 

found in the anterior nares and as such, most people are asymptomatically colonized (Hetem 

et al., 2016; Reygaert, 2013; Wertheim et al., 2005). It has been established that there are three 

types of carriers namely: persistent carriers, intermittent or non-carriers (Hetem et al., 2016; 

Wertheim et al., 2005). Approximately 20% of people are persistent carriers while 30% are 

intermittent carriers (Hetem et al., 2016; Wertheim et al., 2005). Non-carriers are people who 

rarely harbour S. aureus and approximately 50% of healthy people are in this group (Hetem 

et al., 2016; Wertheim et al., 2005). The principal mode of transmission of this organism is by 



 

12  

  

person to person contact and contaminated surfaces or hands as they can survive for long 

periods on such surfaces (Hetem et al., 2016; Reygaert, 2013; Wertheim et al., 2005). Thus, 

basic hygiene practices such as proper and regular cleaning and disinfecting of surfaces have 

been proposed for reducing transmission (Siegel et al., 2006). S. aureus causes mild to severe 

infections. Mild infections include skin and soft tissue infections such as cellulitis, carbuncles 

and scalded skin syndrome. Severe infections include bloodstream infections, meningitis, 

endocarditis, osteomyelitis, pneumonia as well as conditions like toxic shock syndrome and 

food poisoning (Crossley et al., 2009; Ostojić, 2008; Tong et al., 2015). This disease process 

is thought to be facilitated by two mechanisms: first the production of virulence factors or 

toxins which it inherently has and secondly, colonization that causes invasion of the tissue and 

eventual destruction (Reygaert, 2013). In addition to its virulence, S. aureus has an exceptional 

ability to acquire resistance to antibiotics and adapt to its changing environment; factors which 

explain why it has remained a widespread pathogen of medical importance for many centuries 

(Moellering, 2012).   

2.6 Virulence Factors of S. aureus    

S. aureus has several toxin genes and virulence factors such as Panton-Valentine Leukocidin 

(PVL), exfoliative toxins (eta and etb), a range of enterotoxin genes, hemolysins, Protein A 

and other enzymes (Hetem et al., 2016; Reygaert, 2013). These are important factors that 

account for its extraordinary success as a pathogen, as they facilitate adhesion to the surface 

of eukaryotic cells, allow the organism to evade the immune system, colonize and 

subsequently cause disease in its host (Hetem et al., 2016; Reygaert, 2013). Virulence factors 

are found on chromosomes, bacteriophages, plasmids and transposons (Hetem et al., 2016) 

and a number of these factors need to be expressed at a time for pathogenesis to occur.   

Protein A is expressed in almost all S. aureus and contributes to the prevention of opsonisation 

(David & Daum, 2010). It is part of the various surface proteins known as Microbial Surface 



 

13  

  

Components Recognizing Adhesive Matrix molecules (MSCRAMMS) and encoded by the 

spa gene (Crossley et al., 2009). The spa gene has three regions: FC, polymorphic X and C. 

The FC region attaches to the IgG-FC domain of the host to prevent opsonisation, thus, 

hindering phagocytosis by the immune system (Asadollahi et al., 2018).  

Panton-Valentine Leukocidin (PVL) is probably the most extensively studied toxin. This is an 

important toxin known to cause necrotizing pneumoniae which is rapidly fatal (Garnier et al., 

2006). It is also associated with severe infection and recognized as an indicator of community 

associated strains of MRSA (Garnier et al., 2006; Müller-Premru et al., 2005).  PVL is encoded 

by two genes lukS-PV and lukF-PV. These genes act together by binding to specific membrane 

receptors, forming pores in the membrane of the leukocytes and subsequently causing their 

destruction (Chambers & DeLeo, 2009). Hence, PVL is linked with enhanced virulence of S. 

aureus.  

Exfoliative toxins cause exfoliation of the skin epidermis which is followed by secondary 

infections. These toxins are not often found in S. aureus. Four forms ETA, ETB, ETC and 

ETD encoded by the genes eta, etb, etc and etd have been identified, of which A and B are of 

utmost importance in human disease (Crossley et al., 2009). ETA and ETB have been 

associated with neonatal and child staphylococcal scalded skin syndrome (Crossley et al., 

2009) while ETD is associated with wound infections (Dufour et al., 2002).  

Staphylococcal superantigens consist of numerous toxins believed to be associated with 

increased virulence of which the staphylococcal enterotoxins are a part (Holtfreter et al., 2007; 

Tristan et al., 2007). Initially, five serotypes were described, SEA to SEE. However, new 

enterotoxins are constantly being discovered and named alphabetically according to their 

order in which they appear. So far, SEG to SEU has been identified (Blaiotta et al., 2006). 

Enterotoxin A is probably the most studied as it induces a strong proinflammatory response 



 

14  

  

compared to the other toxins (Crossley & Archer, 1997). It has been associated with conditions 

such as septic shock and also with staphylococcal food poisoning, an intoxication caused by 

the release of toxins by the organism before ingestion (Crossley & Archer, 1997). On the other 

hand, SEG, SEI, SELM, SELN, SELO gene clusters are associated with colonization and not 

infection (Dauwalder et al., 2006).  

About 99% of S. aureus produce γ-hemolysins but their roles in pathogenesis is not well 

understood. However, it has been proposed that they play a role in toxic shock syndrome 

pathogenesis together with the Toxic Shock Syndrome Toxin -1 (TSST-1) (Clyne et al., 1988). 

The locus expresses three proteins, namely, HlgA, HlgB and HlgC encoded by the genes hlgA, 

hlgB and hlgC. These proteins assemble to form membrane perforating complexes that are 

toxic to polymorphonuclear cells, monocytes as well as macrophages and causes lysis of red 

blood cells (Hetem et al., 2016).  

2.7 Antimicrobial Resistance in Staphylococcus aureus   

2.7.1 Methicillin Resistance in Staphylococcus aureus  

Methicillin resistance is one of the most significant developments in the history of S. aureus 

evolution. The first drug developed to treat staphylococcal infections was penicillin in the 

1940s by Alexander Fleming. It was thought to be one of the most noteworthy advancements 

in healthcare. However, a few years after its introduction, resistance was developed 

(Deurenberg & Stobberingh, 2008). This resistance was a result of the organism producing an 

enzyme known as beta-lactamase encoded by the blaZ gene that inactivated the drug and 

allowed cell wall synthesis to take place in the organism (Hetem et al., 2016). Methicillin, a 

synthetic penicillin whose structure could not be destroyed by the beta-lactamase enzyme was 

then introduced in the 1960s to treat staphylococcal infections caused by penicillin-resistant 

S. aureus strains (Reygaert, 2013). Unfortunately, resistance was also developed to methicillin 

and down to this day, these strains are resistant to all beta-lactam (β-lactam) antibiotics such 



 

15  

  

as penicillin, cefoxitin, oxacillin and ampicillin as well as to other groups of antibiotics 

including macrolides and aminoglycosides (Chambers, 1997; Chambers & DeLeo, 2009). 

Although methicillin is no more in use and cefoxitin or oxacillin is used for phenotypic testing, 

the term MRSA still exists and is used to refer to S. aureus strains resistant to cefoxitin or 

oxacillin and a wide range of antibiotics. Methicillin resistance is understood to occur when 

Methicillin-Susceptible S. aureus (MSSA) acquires a gene known as mecA by gene transfer. 

This transfer is facilitated by a mobile genetic element called staphylococcal cassette 

chromosome mec (SCCmec). The mecA gene codes for a novel penicillin-binding protein 

PBP2a (Foster, 2004; Hetem et al., 2016). Penicillin-binding protein (PBP) is an essential 

enzyme that catalyzes peptidoglycan production in the cell wall of the bacteria (Farrington, 

2012). All penicillin act by inhibiting the action of this enzyme causing an eventual destruction 

of the organism (Farrington, 2012). PBP2a, an altered form of PBP has low affinity for beta-

lactam, thus, it prevents the beta-lactam ring of the drug from binding to it. This results in 

bacterial cell wall synthesis even when the antibiotic is present and bacterial survival (ECDC, 

2015; Hetem et al., 2016). A divergent homologue of the mecA gene, mecC is also believed 

to confer resistance to methicillin (ECDC, 2015; Hetem et al., 2016). However, mecC codes 

for a different PBP and is usually found in strains from livestock (Hetem et al., 2016).  

Since the late 1960s MRSA has been a major cause of healthcare-associated infections and at 

present is the most widespread resistant pathogen within the hospital (DeLeo et al., 2010). 

MRSA has spread worldwide with varying geographic prevalence (Moellering, 2012). In 

Europe for instance, there are reported variations in MRSA occurrences such as 0.9% in the 

Netherlands and 22.5% in Denmark (ECDC, 2015). However, MRSA prevalence has been 

reported to have decreased from 18.6% to 17.4% between 2011 and 2014 (ECDC, 2015). The 

Netherlands and Scandinavian countries particularly, have successfully reduced MRSA 

probably because of the ‘search and destroy’ policy put in place as well as the implementation 



 

16  

  

of effective infection control measures and restrictive antibiotic use (Hetem et al., 2016; 

Johnson, 2011). A decline is also occurring in the healthcare settings in the US. It has been 

reported that between 2005 and 2011, rates of invasive MRSA dropped by 31% (CDC, 2013). 

In Africa, MRSA isolation rates from 10.5 to 85.5% have been documented among patients 

with SSI (Ahmed, 2012; Dessie et al., 2016; Kahsay et al., 2014; Seni et al., 2013). Elsewhere 

such as India, Iran, Japan and China, rates between 30% and 53% have been reported (Adwan 

et al., 2016; Gu et al., 2015; Sasikumari et al., 2016; Takesue et al., 2017).  In Ghana, attention 

has now been given to MRSA following outbreaks between 2012 and 2015 in some hospitals 

(Amissah et al., 2017; Donkor et al., 2018, 2019). Currently, routine detection of MRSA is 

being conducted in hospitals which was not the case in the past. This probably has contributed 

to surveillance in the country. MRSA prevalence has been reported to be generally low in 

Ghana (Donkor et al., 2019; Egyir et al., 2013). Documented prevalence is between 0.3% and 

34.8% among nasal carriers and patients with infections (Amissah et al., 2015; Dekker et al., 

2016; Egyir et al., 2013; Egyir et al., 2014; Eibach et al., 2017; Odonkor et al., 2012).   

Since MRSA is resistant to both β-lactam and non β-lactam antibiotics, it is considered a threat 

and poses a real challenge in clinical practice, especially, in developing countries where 

antibiotic use is not regulated. There is also the issue of resistant gene transfer to other 

organisms. According to the CDC (2013), about 80,461 invasive MRSA infections occurred 

in 2011 out of which there were 11,285 deaths. Additionally, the European Centre for Disease 

Prevention and Control (ECDC) reported that, annually, 171,200 nosocomial infections are 

associated with MRSA among people in the EU member states, Iceland and Norway. This 

results in an excess hospital cost of €380 million and 5,400 deaths (ECDC, 2009).  

Although MRSA has often been associated with healthcare settings, MRSA has emerged in 

the community among people without any prior exposure to the healthcare setting (Crossley 

et al., 2009; David & Daum, 2010; DeLeo et al., 2010). This strain over time is also showing 



 

17  

  

resistance to numerous antibiotics and moving from the community into healthcare facilities 

(Song et al., 2011) with one reported outbreak occurring in a maternal unit and nursery of a 

US hospital (Bratu et al., 2005). These strains are now even more predominant than 

healthcare-associated infections. Both strains have been shown to have the same virulence 

factors but community acquired strains usually carry a toxin known as Panton Valentine 

Leukocidin (PVL) (Reygaert, 2013). Also, community-associated MRSA tend to be 

susceptible to other classes of antimicrobials apart from β-lactams (David & Daum, 2010) and 

causes more of skin and soft tissue infections (Müller-Premru et al., 2005). Within the 

community, infections have also been associated with MRSA from livestock. These infections 

are believed to be transferred from individuals working with livestock (Cuny et al., 2015). 

MRSA strains are thus classified as healthcare-associated MRSA (HA-MRSA), community-

associated MRSA (CA-MRSA), and livestock-associated MRSA (LA-MRSA), depending on 

whether they are associated with hospitals, communities, or livestock (David & Daum, 2010; 

Gorwitz, 2008; Moellering, 2012).   

2.7.2 Resistance Patterns of S. aureus and MRSA   

Antimicrobials have proven to be useful for the treatment of bacterial infections. However, 

the appearance of resistance has been witnessed in almost all pathogenic bacteria and poses a 

threat (ECDC, 2009). The degree of resistance to commonly used antibiotics varies from 

country to country (ECDC, 2015). This is influenced by the antibiotic use and misuse leading 

to selective pressure (ECDC, 2015). There are a number of antimicrobial drugs used to treat 

staphylococcal infections. These include clindamycin, erythromycin, tetracycline, 

flucloxacillin, cotrimoxazole, linezolid, daptomycin and vancomycin (Naik & Deshpande, 

2011). The main drugs in Ghana are clindamycin and flucloxacillin.   

In India, Kownhar et al. (2008) identified that S. aureus isolates from SSIs were highly 

resistant to gentamicin, penicillin and erythromycin, ampicillin and cefalexin. Similarly, 



 

18  

  

Bhave et al. (2016) found high resistance to penicillin, ciprofloxacin, amoxicillin and 

erythromycin. A study in Shanghai also detected high resistance to almost all antibiotics tested 

especially penicillin, gentamicin, erythromycin, tetracycline and clindamycin (Gu et al., 

2015).    

Across the African continent, resistance patterns of S. aureus to various drugs have been 

documented in many studies (Egyir et al., 2013; Feglo, & Afriyie-Asante, 2014; Kahsay et al., 

2014; Seni et al., 2013). High resistance to penicillin, tetracycline and erythromycin is a 

prominent feature of S. aureus strains in Africa (Schaumburg et al., 2014). This, however, 

may not always be the case as observed in a remote area in central Gabon where majority of 

the S. aureus isolates were susceptible to penicillin and showed a low resistance rate to 

commonly used drugs such as tetracycline and co-trimoxazole (Schaumburg et al., 2011). This 

is the reverse for the study by Kahsay et al. (2014) in Ethiopia where a high proportion of S. 

aureus isolates were found to be resistant to erythromycin, co-trimoxazole, penicillin, 

gentamicin, and amoxicillin while all MRSA isolates detected were 100% resistant to co-

trimoxazole and penicillin. This could infer that the resistance to commonly used antibiotics 

is higher in urban areas than in rural areas. Kesah et al. (2003) also published a study done 

between 1996 and 1997 in Malta and some African hospitals where MRSA isolates were 

sensitive to ciprofloxacin, rifampin and fusidic acid while majority were multi-drug resistant. 

In recent times, Seni et al. (2013) in Uganda also detected that MRSA isolates were sensitive 

to vancomycin and highly resistant to ampicillin and co-trimoxazole.  

Most S. aureus strains circulating in Ghana have shown high resistance to penicillin and 

tetracycline (Amissah et al., 2015; Egyir et al., 2013; Egyir et al., 2014). In a rural area in the 

Ashanti region of Ghana, strains with high resistance to antibiotics frequently used by 

clinicians have been found. These isolates showed resistance particularly to penicillin, 

tetracycline and trimethoprim/sulfamethoxazole (co-trimoxazole) (Dekker et al., 2016). This 



 

19  

  

is a cause for alarm. However, MRSA strains have not been detected in high numbers in these 

areas and these isolated strains are sensitive to other drug used to treat staphylococcal 

infections (Dekker et al., 2016). Similarly, S. aureus from surgical wounds in the Ashanti 

region has been observed to be resistant to penicillin, ampicillin and erythromycin (Feglo & 

Afriyie-Asante, 2014). Conversely, in the Greater-Accra region, MRSA isolates from 

microbiological specimens have been reported with high resistance to penicillin and 

tetracycline as well as to ampicillin and flucloxacillin (Odonkor et al., 2012).  

Vancomycin has been the last resort for treating MRSA infections (Gu et al., 2015). However, 

vancomycin-resistant and vancomycin-intermediate strains have been found (Gardete & 

Tomasz, 2014; Kheder et al., 2012). This makes serious infections caused by MRSA even 

more difficult to treat because there are a limited number of antimicrobial agents that could 

be used. Most of the studies mentioned above have documented sensitivity to vancomycin. In 

Ghana, high susceptibility to vancomycin may be attributable to the fact that these drugs are 

mostly unavailable and as such not frequently used (Labi et al., 2018).  

2.8 Typing Methods for S. aureus   

Typing is an essential tool for studying genetic diversity and determining the spread of 

organisms in order to develop effective surveillance systems for control (Deurenberg & 

Stobberingh, 2008). Many typing methods can be used to distinguish bacterial strains based 

on the genotypic characteristics and the occurrence of specific genes or genetic elements. 

These methods may be band or sequence-based. For S. aureus, the current methods used are 

Pulsed Field Gel Electrophoresis (PFGE), Staphylococcal Chromosome Cassette mec 

(SCCmec) typing, Staphylococcus Protein A (spa) typing and Multilocus Sequence Typing 

(MLST). These methods are widely used and vary in reproducibility, speed, ease 

discriminatory capacity as well as cost (Deurenberg & Stobberingh, 2008; Hetem et al., 2016). 

However, the type of method to use depends on the problem to be addressed (O’Hara et al., 



 

20  

  

2016; Sabat et al., 2013). Methods with high discriminatory power are best used when 

characterizing for epidemiological investigation (O’Hara et al., 2016; Rodriguez et al., 2015).  

2.8.1 Pulsed Field Gel Electrophoresis (PFGE)  

PFGE has been considered the gold standard and replaced phage typing. It is used widely for 

investigation of outbreaks and local epidemiological studies (Gu et al., 2015). This method 

however, is limited by the fact that it is time-consuming, comparing results from different 

laboratories is difficult, requires specialized equipment and is more expensive (Hetem et al., 

2016). PFGE has gradually been replaced by MLST and spa typing.  

2.8.2 Spa Typing  

This is a single-locus typing method based on variations in the sequences at the polymorphic 

X region of the spa gene which encodes Staphylococcus Protein A (Chambers & DeLeo, 

2009). The X-region is made up of variable numbers of 24bp repeat sequences. Each 24bp 

repeat sequence is assigned a unique code. A combination of the different repeats gives the 

repeat succession of each isolate and determines its spa type. Diversity in the spa gene is often 

attributed to deletions, point mutations and duplications of the repeats (Chambers & DeLeo, 

2009; Deurenberg & Stobberingh, 2008). This method has been used to study S. aureus and 

MRSA outbreaks as well as their molecular evolution (Chambers & DeLeo, 2009).   

One advantage of this method is the availability of a software tool known as Ridom 

StaphTyper that provides easy sequence analysis. This software helps in spa type 

determination and provides a universal spa server database (http://spaserver.ridom.de) that 

allows for comparison of data across laboratories (Deurenberg & Stobberingh, 2008). 

Recently, a free accessible tool on the Centre for genomic epidemiology web page 

(http://www.genomicepidemiology.org/) has also been developed. Other advantages of this 

method over the others are that it is reproducible, easy to interpret and less expensive since it 

http://spaserver.ridom.de/
http://spaserver.ridom.de/
http://www.genomicepidemiology.org/
http://www.genomicepidemiology.org/
http://www.genomicepidemiology.org/


 

21  

  

involves only a single locus (Hetem et al., 2016). Some though have questioned this method 

because of the use of a single locus.  

2.8.3 Multi-Locus Sequence Typing (MLST)  

MLST has been described as a great tool for studying the long periods of evolution in S. aureus 

(Chambers & DeLeo, 2009). This method is based on variations in the sequences of seven 

house-keeping genes arcC, aroE, glpF, gmk, pta, tpi and yqiL (Deurenberg & Stobberingh, 

2008). These genes are essential for bacterial survival thus present in all S. aureus. Each 

variation in the sequence of the gene is assigned a number, yielding a 7-numbered code that 

is specific for each sequence type (ST) and bacterial isolate. Since the housekeeping genes are 

not subject to a lot of change and selective forces, they provide more reliable information. 

However, this method is considered not suitable for routine infection control anymore as it is 

more expensive, time consuming and laborious (Strommenger et al., 2006). Yet, it is easy to 

compare or exchange the data generated internationally (Hetem et al., 2016).  MLST and spa 

typing are most often used in combination with software-based clustering algorithms like 

BURST (Based Upon Related Sequence Types) and BURP (Based Upon Repeat Pattern) to 

place related isolates into clonal complexes (CCs) (O’Hara et al., 2016).  

2.8.4 SCCmec Typing   

SCCmec typing is currently the primary typing method for MRSA characterization. This is a 

PCR based method used to determine the structures of the mobile genetic element SCCmec 

on which the mecA gene is borne (Deurenberg & Stobberingh, 2008). There are three protocols 

by which the SCCmec types are identified (Deurenberg & Stobberingh, 2008).    

2.8.5 Whole Genome Sequencing (WGS)  

Whole genome sequencing is not a widely used tool for molecular characterization of bacterial 

strains because of its cost. However, recent advances in sequencing have made it more 

affordable and accessible. This tool allows for detailed analysis of entire genomes of 



 

22  

  

organisms (Kwong et al., 2015). WGS resulted from a revolution in sequencing technologies 

following the Human Genome Project. Advances were initially focused on improving Sanger 

sequencing developed in 1977 by Sanger and others. However, more efficient tools were later 

sort after which led to the establishment of ‘Shotgun sequencing’ (Kwong et al., 2015). 

However, this was laborious for sequencing whole genomes. Later advances in this 

technology led to the development of ‘whole-genome shotgun sequencing’ (Kwong et al., 

2015).  In recent times, sequencing technologies such as Next-Generation Sequencing (NGS) 

with easy and efficient library preparation protocols as well as simple benchtop systems have 

been developed (Kwong et al., 2015). This has made WGS more affordable, efficient and 

accessible for clinical and research applications (Kwong et al., 2015).   

NGS methods generally have three steps: library preparation, amplification and sequencing. 

A number of library preparation kits are available that offer different protocols for sequencing 

both bacterial and viral pathogens, however, sequencing principles remain the same (Illumina, 

2015). As opposed to Sanger sequencing which typically produces longer reads, NGS 

produces millions of small fragments of DNA in parallel (Behjati & Tarpey, 2013). The 

Roche/454, ABI/SOLiD, and Solexa/Illumina are the main NGS sequencing platforms, 

however, the Illumina is by far the most widely used technology (Hodzic et al., 2017). An 

additional and important part of sequencing is data analysis. Post-sequence data analysis can 

be done with commercially available or open source tools some of which have been developed 

for people with limited bioinformatics (Kwong et al., 2015). Analysis basically involves 

piecing together the fragments and mapping them to a reference genome from which a 

numerous analyses are possible (Behjati & Tarpey, 2013; Illumina, 2015).  

There are a number of applications of WGS currently, which include identification of 

organisms, typing of bacterial strains, and detection of resistance and virulence factors 

(Kwong et al., 2015). Documented studies have used WGS to investigate outbreaks (Harris et 



 

23  

  

al., 2013; Kong et al., 2016), predict antimicrobial resistance (Gordon et al., 2014), type 

bacterial strains (Salipante et al., 2015) and to describe the transmission of MRSA (Harris et 

al., 2013).  

Using this tool is considered better than other methods for typing because it gives a better 

resolution of bacterial isolates (Salipante et al., 2015). It also outperforms PFGE when it 

comes to discriminatory power and reproducibility (Salipante et al., 2015).WGS is however 

limited by the fact that analyses will depend on  the quality of the genome assembly or 

sequencing and also on the quality and selection of the reference genome (Kwong et al., 2015). 

In addition, even though WGS data provides an extensive genomic information, it does not 

necessarily provide information about gene expression or transcription (Kwong et al., 2015). 

It has been proposed that other known methods of typing will probably be replaced by WGS 

as it represents the acme for characterisation of strains and epidemiological analyses (Kwong 

et al., 2015).  

2.9 Molecular Epidemiology and Genetic diversity of Staphylococcus aureus   

MRSA strains are biologically diverse and have been shown to vary from one geographic area 

to the other (Gorwitz, 2008) with some strains becoming dominant within certain areas and 

others globally distributed. In Eastern Australia, the common sequence type is ST30 while in 

Europe the major sequence type is ST80, and in Taiwan ST59 (Chambers & DeLeo, 2009).  

Even so, in Singapore all these strains have been reported possibly because there are millions 

of people visiting every year as it serves as an international travel hub (Hsu et al., 2005). 

Asadollahi et al. (2018) conducted a review to determine the predominant spa types in the 

world among clinical isolates. Results of the study indicate that the most prevailing spa type 

in Europe mainly UK and Germany is t032 followed by t008. This spa type t008 was also 

reported as the most prevalent in the USA and Canada. In Asia, the predominant spa type was 



 

24  

  

t030 followed by t037. Spa type t030 was also reported as the fifth most common type in Iran. 

In America, t002 was reported as the second most common spa type. This review also 

identified that there were sustained associations between spa type and sequence types. ST22 

always associated with t032 regardless of the continent while ST8 and ST247 associated with 

t008. Some associated types were detected including: t002 with ST5, t030 with ST239, also 

ST22 and t037 with ST239. Another research aimed at determining the distribution of clones 

causing S. aureus invasive infections in Europe found that MSSA was more diverse than 

MRSA and among MSSA, the strains t002 (ST5) and t084 (ST15/18) were the most 

predominant. On the other hand, t032 (ST22) was predominant among MRSA isolates 

(Grundmann et al., 2010).   

The PVL toxin is an important virulence factor and its carriage has been closely linked to 

infections caused by CA-MRSA (David & Daum, 2010). In Europe, PVL has been associated 

with CA-MRSA with a prevalence of less than 2% while in North America the prevalence is 

high (Rasigade et al., 2014). Contradictory to this is PVL-positive strains found on the African 

continent. These have been detected in high numbers with prevalence ranging from 17% to 

74% and most often associated with MSSA. (Breurec et al., 2011; Egyir et al., 2014; Rasigade 

et al., 2014). It has been suggested that PVL-negative ST5 which is a paediatric clone and 

ST22 known as Barnim epidemic strain/EMRSA-15 were widespread for many years before 

PVL-positive ST5 and PVL-positive ST22 existed in Slovenia and Bavaria respectively 

(Linde et al., 2005; Müller-Premru et al., 2005). This worldwide epidemic strain ST5 was 

detected in skin and soft tissue infections associated with spa type t002 and PVL (Müller-

Premru et al., 2005). This was the first study reporting PVL in ST5 in Europe.   

ST152 and ST88 are strains often isolated from patients with a history of travel. ST152 has 

been identified  as a pandemic clone in Macedonia and also in Slovenia (Monecke et al., 2007; 

Müller-Premru et al., 2005) suggesting that they are mainly distributed in Balkans (Monecke 



 

25  

  

et al., 2007). In Slovenia, it was detected in a team of footballers with severe skin and soft 

tissue infections caused by CA-MRSA. It should be noted that these footballers had no 

exposure to the healthcare setting (Müller-Premru et al., 2005). This pandemic clone has also 

been associated with PVL-positive CA-MRSA in Central Europe, the Balkan and Switzerland 

(Monecke et al., 2007; Ruimy et al., 2008).   

In Middle Eastern countries including Gaza, CA-MRSA strains are mostly PVL-negative 

ST22. However, in Egypt and Jordan, PVL-positive ST80 known as the European clone is 

causing the CA-MRSA epidemic found there (Rasigade et al., 2014).   

2.9.1 Molecular Epidemiology and Genetic Diversity of Staphylococcus aureus in Africa 

It is not until recently that efforts have been made to describe the molecular epidemiology of 

S. aureus strains in Africa. The majority of African countries including Ghana are widely 

dominated by spa type t084 and t355 (Amissah et al., 2015; Breurec et al., 2011; Dekker et 

al., 2016; Egyir et al., 2014; Schaumburg et al., 2014). ST239/241-MRSA-I/III/IV which is a 

Hungarian/ Brazilian clone is also prevailing in the African continent (Schaumburg et al., 

2014). Other clones such as the well-known ST239/241-III pandemic clone (Deurenberg & 

Stobberingh, 2008) has been found among isolates from a multi-centre study across five towns 

in Madagascar, Morocco, Niger, Cameroon and Senegal (Breurec et al., 2011). ST88-MRSA- 

III/IV occasionally found in certain countries except Far East Asia also seems to be linked 

with Africa and has also been reported in Madagascar (Breurec et al., 2011; Schaumburg et 

al., 2014). Likewise, ST80-MRSA-IV a PVL- positive CA-MRSA clone found mainly in 

Europe but rarely found in sub-Saharan Africa is predominant in Maghreb (Schaumburg et 

al., 2014).  

In addition, ST5-MSSA and ST5-MRSA have been observed to be widespread in Central and 

West Africa (Schaumburg et al., 2014). Schaumburg et al., (2014) further suggested that 



 

26  

  

considering that both strains are found in the same geographic area, ST5-MRSA may have 

evolved from ST5-MSSA acquiring SCCmec. However, this may not be the case for 

ST8MRSA and ST8-MSSA since they have different geographic locations. ST8-MSSA from 

Maghreb and ST8-MRSA from Central and West Africa.   

Some publications have appeared in recent years documenting the diversity of S. aureus in 

Ghana. The prevalent spa types in Ghana are t355 and t084 (Amissah et al., 2015; Egyir et al., 

2014; Eibach et al., 2017). Other spa types that have also been detected include: t314 and t311 

(Amissah et al., 2015, 2017; Dekker et al., 2016; Donkor et al., 2019; Egyir et al., 2014; Eibach 

et al., 2017). The common pandemic clone ST152 has also been identified in Ghana as 

predominant, with the above mentioned studies documenting this as well as that of Kpeli et 

al. (2016). Many other sequence types are found to be abundant such as ST15, ST121 and ST5 

associated with MSSA (Dekker et al., 2016; Egyir et al., 2014). Donkor et al. (2018) reported 

ST15 as a possible cause of the outbreak at the paediatric emergency ward of Korle-Bu 

Teaching Hospital. A study by Amissah et al. (2017) also reported high numbers of MRSA 

belonging to ST250 in a burn unit at KBTH although this was not PVL- positive. In this same 

study, four new spa types were identified: ST3248, ST3249, ST3250 and ST3251. These S. 

aureus isolates have also been associated with high PVL positivity which buttresses the point 

that Africa is a PVL endemic region (Schaumburg et al., 2014).  

 

 

 

 



 

27  

  

CHAPTER THREE  

3.0 MATERIALS AND METHODS   

3.1 Study Design   

This was a hospital-based cross-sectional study conducted over a 6-month period from June 

to November 2018 as part of a larger study entitled “Prevalence of MRSA/ESBL Producing 

Bacteria Associated with Surgical Site Infections in a Military and Civilian Hospital in Ghana” 

at Noguchi Memorial Institute for Medical Research (NMIMR). Patients seeking healthcare 

who had undergone surgery with infections at the site of incision were recruited into the study 

at the selected departments of the Korle-Bu Teaching Hospital and 37-Military hospital.   

3.2 Study Area  

The study was conducted at the 37-Military Hospital and Korle-Bu Teaching Hospital 

(KBTH) in Accra.   

3.2.1 37-Military Hospital  

The 37-Military hospital is the largest military hospital and also a specialist hospital located 

in Accra, Ghana. It has about 400 beds, a twenty four (24 hr) accident and emergency 

department and pharmacy (“The Electives Network: 37 Military Hospital,” n.d.). It caters for 

both military and civilian patients as such, it sees a diverse number of patients. This site was 

included in the study because it also serves as a referral hospital and thus, receives referral 

cases. Additionally, the funding body [Global Emerging Infectious Surveillance (GEIS)] 

supports mainly military surveillance research.  

The study was conducted at the following departments in the hospital: surgical OPD, Ghandi 

ward, Tamakloe ward, trauma and surgical emergency unit and orthopaedic ward. The surgical 

OPD sees patients who had surgeries and have been discharged from the hospital. These 

patients return to the hospital about twice a week for wound dressings and may be recruited 



 

28  

  

into the study. The Ghandi ward is recovery ward for female patients after surgery while the 

Tamakloe ward is a recovery ward of both male and female surgical patients. On the other 

hand, the trauma and surgical emergency unit is the first point of call for patients involved in 

accidents. A few of these patients who undergo surgery and are discharged often report to the 

unit when they have infections.    

3.2.2 Korle-Bu Teaching Hospital  

Korle-Bu teaching hospital is the largest hospital and leading referral hospital in Ghana. It has 

a hospital bed capacity of 2000 as well as 17 clinical and diagnostic departments which 

comprise obstetrics and gynaecology, surgical/medical emergency, surgery, and an accident 

centre (“About us – Brief History – Korle-Bu Teaching Hospital,” n.d.). An average of 1,500 

people visit the hospital each day and there are about 250 admissions (“About us – Brief 

History – Korle-Bu Teaching Hospital,” n.d.). Apart from being a referral hospital, it is also a 

teaching hospital which trains medical doctors and a wide range of health professionals. There 

is also a 42-bed polyclinic that provides care for residents in the community as well as other 

parts of Accra (“About us – Brief History – Korle-Bu Teaching Hospital,” n.d.). Initially, it 

was set-up to be a service delivery facility within the community but has developed beyond 

that. This site was included in the study because it receives lot of referral cases from other 

regions of the country in addition to referrals from all parts of the city. Most importantly, this 

is the hospital where most complicated surgeries are performed.  

Patients recruited from this health care facility were from the following departments: 

obstetrics and gynaecology, surgical, orthopaedics, maternity, neurosurgery and paediatrics. 

3.3 Ethical Clearance   

Ethical approval was obtained from the Noguchi Memorial Institute Institutional Review 

Board, Korle-Bu Institutional Review Board, 37-Military Hospital Institutional Review Board 

and the Naval Medical Research Centre Institutional Review Board.   



 

29  

  

Approvals were also sought from the various departments of the Korle-Bu Teaching Hospital 

and the 37-Military hospital where the study was conducted.  

3.4 Recruitment of Study Participants   

Out-patients reporting to the hospital for wound dressing after surgery, as well as in-patients 

who had undergone surgery with infections and fit the eligibility criteria were recruited into 

the study. Participants were either given consent forms to read and sign or had the consent 

forms read and explained to them, then the forms were signed in the presence of a witness. 

All participants below the age of eighteen were given child assent forms and parental consent 

was also sought from their guardians. Participants were then given a study ID to maintain 

confidentiality. The field staff completed the case report forms using the patient folder 

information and patient’s answers. Demographic data such as age, sex, as well as type of 

surgical operation, type of infection, co-morbidities, period of hospitalization and 

antimicrobial therapy were also collected.  

3.4.1 Inclusion Criteria   

Selected participants comprised patients who had undergone surgery and had developed 

infections with purulent discharge at the site of incision. The infection should have been 

diagnosed by a physician and met the CDC criteria for classification as a surgical site infection 

(CDC, 2015; Mangram et al., 1999).   

3.4.1.1 CDC Criteria  

The CDC requires that for a patient to meet the criteria, the patient should have had a National 

Healthcare Safety Network (NHSN) operative procedure. NHSN defines an operative 

procedure as one that takes in an operating room where at least one incision is made through 

the skin or mucous membrane or reoperation through an incision that was left open during a 

prior operative procedure. The period for infection should be 30 or 90 days after surgery 

depending on the procedure performed and the depth of the infection (CDC, 2015; Mangram 



 

30  

  

et al., 1999). NHSN provides a list of operative procedures that should be followed for 30 or 

90 days. This table can be found in Appendix IX.  

3.4.2 Exclusion Criteria   

Patients who had undergone surgery, had developed infections but not within the CDC 

classification as a surgical site infection were excluded.   

3.5 Sample Collection   

A trained study nurse or a clinician aseptically collected either a wound swab, fluid or aspirate 

from each patient using either a sterile cotton-tipped applicator or a syringe respectively. The 

wound was first cleaned with normal saline to reduce the skin flora before the samples were 

taken. For wound swabs, a part of the handle of the applicator stick was broken off and the 

rest of the stick was placed in a 15ml falcon tube containing 10ml of thioglycollate broth. 

Aspirates and fluids were dispensed into the tube with the thioglycollate broth. The samples 

were placed in an ice chest and transported at room temperature to NMIMR within 24 hours 

to maximize the chance of isolating the causative organism.  

3.6 Laboratory Procedures  

3.6.1 Sample Processing   

Upon arrival at the Naval Medical Research Unit three (NAMRU-3) laboratory at NMIMR, 

the sample IDs were entered into a sample receiving book and inspected to make sure the right 

samples were collected. The samples were then incubated aerobically at 37°C for 18-24 hours. 

After incubation, the samples were cultured on mannitol salt agar, blood agar and then 

incubated aerobically for another 18-24 hours.  

3.6.2 Identification of Bacterial Isolates   

Staphylococcus spp isolates were identified using colonial morphology, Gram stain 

characteristics, catalase test and coagulase test. Using colonial morphology, Staphylococcus 



 

31  

  

spp were identified by 2mm white or cream-coloured to golden beta-hemolytic colonies on 

blood agar plate, and 2-3mm yellow colonies on mannitol salt agar plate (Forbes et al., 2007; 

Maza et al., 2004).   

3.6.2.1 Gram Stain  

Using a sterile1mm loop, one to two colonies of freshly cultured bacteria was placed on a 

clean slide containing a drop of saline. The colonies were emulsified in the saline and allowed 

to air-dry. All slides were heat fixed prior to Gram staining. Gram stain was performed using 

the BD BBL Gram Stain kit (Becton Dickinson, Australia). Staining was done by adding 

crystal violet for one minute after which the slide was washed with water. Subsequently, the 

slide was flooded with iodine for another minute and washed off. The slide was then 

decolourized with a decolourizer then immediately washed off with water. Finally, Safranin 

was added for another minute and washed off. The slides were allowed to air-dry and viewed 

under the oil immersion lens (100X) of the microscope. Gram-positive cocci in clusters were 

observed.  

3.6.2.2 Catalase Testing  

The catalase test was performed using a catalase reagent dropper (Becton Dickinson, 

Australia). A drop of the reagent was placed on a clean slide and a colony of the bacteria was 

placed in the reagent. A catalase-positive test was observed as a fizzing reaction. This 

indicated a conversion of hydrogen peroxide to water and oxygen.  

3.6.2.3 Coagulase Testing  

A coagulase test was then used to differentiate S. aureus isolates from other staphylococci.  

Coagulase testing was performed using the BD BBL Coagulase rabbit plasma (Becton 

Dickinson, Australia). One millilitre (1ml) of rabbit plasma was placed in a sterile tube and a 

few colonies of the test bacteria was added to it. The tubes were incubated aerobically at 37°C 



 

32  

  

for four hours. A coagulase-positive reaction was observed by formation of a clot. For isolates 

that showed negative after four hours, they were re-incubated for up to 24 hours to confirm 

they were negative. A coagulase-positive standard strain Staphylococcus aureus ATTC 25923 

and a coagulase-negative standard strain Staphylococcus epidermidis ATCC 12228 was used 

as a quality control.  

3.6.3 Confirmation of Isolates Identified  

All Staphylococcus spp isolates were confirmed using the Matrix Assisted Laser 

Desorption/Ionization (MALDI) Biotyper (Bruker, USA).  

The MALDI machine has three main components which are the ionization source, the analyser 

and detector as well as an inlet for sample loading. Once the sample is loaded, it is ionized by 

matrix-assisted laser desorption ionization (MALDI) method. The machine then uses a 

database of known organism to match the isolate on the target plate while providing a 

matching score. Scores >2 are considered as very good matches.  

Samples were prepared by placing 1-2 colonies of Staphylococcus spp on a stainless-steel 

plate known as the target plate using a 1µ loop and allowed to air-dry. One microliter (1µl) of 

formic acid was then added and allowed to air-dry. Finally one microliter (1µl) of matrix was 

added, allowed to air-dry and the target plate was then placed in the machine. The matrix is 

an organic, energy-absorbent compound which facilities the ionization process. It works to 

absorb energy in the form of ultraviolet light and transfers to the larger sample molecules.  

Once inside the MALDI machine, the target is brought to a vacuum and struck by a pulsed 

ultraviolet laser. The matrix absorbs the laser energy and then it is gradually removed from 

the surface of the sample carrying along the analyte molecules of the sample with it into a 

gaseous phase. The process of removal of known as ablation. During the ablation process, the 

analyte molecules are ionized by proton transfer from the nearby matrix molecules. The gas 



 

33  

  

phase analytes ions are then analysed by the Time of Flight (TOF) spectrometer. The TOF 

spectrometer accelerates the gas phase ions in a high voltage electric field which imparts a 

constant amount energy and this causes the smallest ions to travel the fastest. The TOF 

detector then records the time it takes for the different groups of ions to travel a certain 

distance. The spectrometer is then able to calculate and record the mass to charge ratio of each 

ion it detects. The TOF information is used to generate a characteristic spectrum for the 

analytes in the sample. This is known as the Peptide Mass Fingerprint (PMF). These peptides 

are proteins unique to the organism that are cleaved into smaller fragments. The 

microorganism is identified by comparing the PMF of the unknown organism with the PMF 

of the known organism in the systems database. This allows for identification of the 

microorganism to the genus and in many instances the species level (Singhal et al., 2003). It 

took just a few minutes for the machine to generate these results.    

3.6.4 Antimicrobial Susceptibility Testing  

Antimicrobial testing was performed using the Kirby-Bauer Disk diffusion method. An 

inoculum equivalent to 0.5 McFarland standard was prepared using physiological saline and 

the turbidity was measured with a Nephelometer. The inoculum was streaked on Mueller 

Hinton agar using a sterile cotton tipped applicator after which the antibiotic impregnated 

disks were placed on the agar aseptically and incubated aerobically at 37°C for 18-24hrs.  

Antibiotics recommended by Clinical and Laboratory Standard Institute (CLSI) and The 

European Committee on Antimicrobial Susceptibility Testing (EUCAST) were used for 

testing. The antibiotics tested included: Penicillin (1Unit), Cefoxitin (30µg), Tetracycline 

(30µg), Clindamycin (2µg), Erythromycin (15µg), Gentamicin (10µg), Rifampicin (5µg), 

Trimethoprim/Sulfamethoxazole (1.25/23.75µg), Linezolid (10µg), Norfloxacin (10µg) and 

Vancomycin E-test strips. The zones of inhibition were measured with a digital calliper and 

zone sizes interpreted using the CLSI (2018) guideline. For antibiotics [Penicillin (1Unit), 



 

34  

  

Linezolid (10µg), and Rifampicin (5µg)] whose concentrations could not be found in the CLSI 

guideline, the EUCAST 2018 guideline was used. All isolates with Cefoxitin zone sizes 

≤21mm were considered presumptive MRSA. The S. aureus standard strain ATCC 25923 was 

used as a quality control for all antimicrobial sensitivity testing.  

3.7 Molecular Techniques   

3.7.1 DNA Extraction   

Crude DNA from was extracted from S. aureus isolates as described by Dashti et al. (2009) 

with slight modifications. Briefly, 2-6 pure colonies from an overnight culture were suspended 

in a 1.5 ml Eppendorf tube containing two hundred microliters (200µl) of nuclease-free water. 

The tube was placed on a heating block set at 100°C and allowed to boil for 10 minutes. The 

suspension was then centrifuged for 5 minutes at a speed of 13,000rpm to separate the pellet 

and supernatant. Finally the supernatant was pipetted into a new 1.5 ml Eppendorf tube leaving 

behind the pellet. All extracts were stored at -20°C until further testing.  

3.7.2 Polymerase Chain Reaction   

A multiplex PCR was performed to detect the spa, pvl and mecA genes following the Larsen 

et al. (2008) protocol with slight modifications. A primer mix containing primers for all genes 

mentioned in Table 3.1 was prepared prior to master mix preparation. The concentration of 

each primer (forward and reverse) was as follows 0.18µM for spa, 1µM for pvl and 0.45µM 

for mecA. The primer sequences are listed in Table 3.1. To make a (1ml) primer mix 

containing the above concentrations, a volume of each 10µM forward and reverse primer was 

picked and an amount of water was added. The volumes for each primer are listed in Table 

3.2. Each PCR tube contained a 25µl reaction volume consisting of 12.5µl Multiplex PCR 

Mastermix (Qiagen, Germany), 2.5µl RNase-free water (Qiagen, Germany), primer mix of 

8µl and a DNA template of 2µl. DNA amplification was performed using an Eppendorf 

Mastercycler x50s with the following cycling conditions: initial denaturation at 94˚C for 



 

35  

  

15mins, 30 cycles of 94˚C for 30sec, 57˚C for 1 min, 72˚C for 1min followed by a final 

extension at 72˚C for 10mins. A positive control showing all three genes and a negative control 

(nuclease-free water) were included to ensure the results were accurate. 

Table 3.1: Primer Sequences used in the Study  

Primer  Primer sequences   Description   References  

spa   F: 5'-TAAAGACGATCCTTCGGTGAGC-3’ 

R: 5'-CAGCAGTAGTGCCGTTTGCTT-3’  

To detect 

spa gene (S. 

aureus 

specific)  

(Larsen et 

al.,  

2008)  

pvl   F: 5'-GCTGGACAAA   

ACTTCTTGGAATAT-3’  

R: 5'-

GATAGGACACCAATAAATTCTGGATTG-

3’  

To detect 

Panton  

Valentine  

Leukocidin  

(virulence 

factor)  

(Larsen et 

al.,  

2008)  

mecA  F: 5'-TCCAGATTACA ACTTCACCAGG-3’ 

R: 5'-CCACTTCATATCTTGTAACG-3’  

To   detect  

methicillin 

resistance  

(Larsen et 

al.,  

2008)  

  

  

Table 3.2: One millilitre (1ml) Primer mix preparation  

Primer   Forward   Reverse  

mecA (10uM)  45µl    45µl  

spa (10uM)  18µl   18µl  

pvl  (10uM)  100µl   100µl  

Nuclease free water    674µl  

Total  volume   1000µl  

  

3.7.3 Gel Electrophoresis  

A 2% w/v agarose gel was used to analyse the amplified products. The expected band sizes 

for the various genes were: spa (variable region- 200-600bp), mecA (162bp), and pvl (80bp)  



 

36  

  

3.7.3.1 Gel Preparation  

The gel was prepared by first weighing 1.5g of agarose powder and adding to 75 ml of Tris 

acetate-EDTA (TAE) buffer which had been measured into a beaker. The TAE buffer and 

agarose powder mixture was microwaved for at least 2 mins or until the mixture was clear. 

The resultant mixture was allowed to cool. SYBR Safe DNA gel stain (10,000X concentrate) 

(Invitrogen, USA) of 7.5µl was added and mixed thoroughly. The mixture was then poured 

into a casting tray with combs inserted and left for at least 30 mins to solidify after which the 

combs were removed.   

3.7.3.2 Gel loading and Running  

The casting tray containing the gel was transferred into the gel tank and the tank was filled 

with 1X TAE Buffer. The wells were loaded with 5ul of a 100bp ladder, 7ul of each sample 

as well as a positive and negative control. A blue/orange 6x loading dye (Promega, USA) was 

used to load the PCR products. The gel was run at 80 Volts for at least one (1) hr. 

Subsequently, the gels were visualised using the LED Illuminator (BT Lab Systems) and 

photographed using the UVP BioDoc-It Imaging system (Analytik Jena, Germany).  

3.7.4 DNA Extraction for Sequencing  

For sequencing purposes, DNA was re-extracted using the Lucigen extraction kit following 

the manufacturer’s instructions with slight modifications. Here, the bacterial colonies were 

picked from the culture plate with an inoculation loop and suspended in 300µl of 1X 

Phosphate Buffered Saline (PBS). The suspension was vortexed and centrifuged at 10,000 

rpm for 5 mins. Two hundred and seventy-five microliters of the supernatant was discarded 

leaving behind the pellet and approximately 25µl of the supernatant. The pellet was then 

resuspended by vortexing for 10 seconds. One microliter of Proteinase K was diluted into 

300µl of tissue and cell lysis solution and added to each sample. The tube was vortexed to 

thoroughly mix its contents and incubated at 65°C for 30 minutes on a rocking platform. After 



 

37  

  

incubation, the samples were cooled to 37°C and 1µl of 5mg/ml RNase A was added. The 

mixture was incubated again at 37°C for 1 hour and then placed on ice for 3-5 minutes. 

Subsequently, 150µl of MPC Protein Precipitation reagent was added and vortexed for 10 

seconds. The tube was then centrifuged at 10,000xg for 10 minutes at a temperature of 4°C. 

For pellets that were clear, small or loose, an additional 25µl of MPC Protein Precipitation 

reagent added and centrifuged again. The supernatant was transferred into a new 1.5ml tube 

and 500µl of isopropanol was added to it. Afterwards, the tube was inverted 30-40 times and 

centrifuged for 10 minutes at a temperature of 4°C. Without disturbing the pellet, the 

isopropanol was pipetted out and the pellet was rinsed twice with 70% ethanol by adding and 

pouring out while still ensuring the pellet is not lost. Finally, the pellet was resuspended in 

35µl of TE buffer. DNA extracts were stored in -20℃ until further use.  

3.7.5 Whole Genome Sequencing   

Sequencing was done using the Nextera DNA Flex Library preparation kit and sequenced with 

the Mi-Seq Illumina sequencer. The library preparation was done according to the 

manufacturer’s instruction.  

The extracted DNA was quantified using a Qubit double strand (ds) high