MALARIA AND BACTERIAL CO-INFECTIONS: A STUDY AMONG CHILDREN PRESENTING WITH FEBRILE ILLNESSES IN ACCRA. BY RAYMOND BEDU AFFRIM (10397193) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON, IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL MICROBIOLOGY DEGREE JUNE 2015 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I hereby declare that this thesis is my original work and has not been presented for a degree in any other institution. I have duly acknowledged references made to other authors’ work in the reference section of the thesis. Student: Signature:……………………………………………Date………/.………/…..…… Mr. Raymond Bedu Affrim Supervisors: Signature:…………………………………………… Date………/.………/…..…… Rev. Professor. Patrick Ferdinand Ayeh-Kumi Department of Microbiology, University of Ghana School of Biomedical and Allied Health Sciences, College of Health Sciences. Signature:…………………………………………… Date………/.………/…..…… Professor Ben Gyan Department of Immunology, Noguchi Memorial Institute of Medical Research, University of Ghana. University of Ghana http://ugspace.ug.edu.gh ii DEDICATION To the memory of my mother Miss Florence Akua Kofituo. Thank you for the gift of education, advice and care which has brought me this far. To all children in deprived settings burdened with malaria and bacterial co-infections who are unable to seek quality healthcare. You are the inspiration for such studies; there is definitely light at the end of the tunnel. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENT I give thanks unto the Almighty God for giving me strength and wisdom to carry out this project. My profound appreciation goes to my supervisory team, Professor Ben Gyan, Rev. Professor Patrick Ayeh-Kumi, Dr. Simon Attah and Dr. Patience Tetteh-Quacoo, I say a big thank you for your constructive criticisms and encouragements without which this work would not have been a success. I am grateful to Professor Dorothy Yeboah Manu, Head of Microbiology Department of NMIMR for giving me the permission to use laboratory facilities in the Department for my bench work. To the Head of Microbiology Department - SBAHS, Dr Theophilus Adiku for your constructive guidance and advice for the successful completion of this work. Many thanks go to Professor Eric Sampane- Donkor, Dr. Japhet Opintan, Dr. Korang Larbi, Rita Ofosu Agyemang and all the staff and students of the Medical Microbiology Department, School of Biomedical and Allied Health Sciences for their moral support and encouragement. I acknowledge with deep appreciation, the indispensable help from my nephew Mr. Francis Dzidefo Krampa. Special thanks to Mr. Lorenzo Arkyeh, Christian Bonsu, Emelia Danso, Elias Asuming-Brempong and all staff of the Microbiology Department, NMIMR for being accommodative and lending their hands of support. I wish to express my heartfelt gratitude to all my siblings, and especially my children for their immense contribution towards my education. I am indebted to Mr. Thomas Dankwah, Richael Mills, Dominic Edu and John Nyarko of the Central Laboratory, KBTH for their assistance in lab analysis. Finally, am grateful to the NMIMR for allowing me to access the Postgraduate Fund in order to research further into the immunologic aspects of this study. University of Ghana http://ugspace.ug.edu.gh iv TABLE OF CONTENTS DECLARATION ............................................................................................................. i DEDICATION ................................................................................................................ii ACKNOWLEDGEMENT ............................................................................................ iii LIST OF TABLES ........................................................................................................vii LIST OF FIGURES ......................................................................................................vii LIST OF ABBREVIATIONS ..................................................................................... viii ABSTRACT ................................................................................................................... ix CHAPTER ONE ............................................................................................................. 1 INTRODUCTION .......................................................................................................... 1 1.1 BACKGROUND .............................................................................................. 1 1.2 PROBLEM STATEMENT ................................................................................... 3 1.3 JUSTIFICATION / RATIONALE ........................................................................ 4 1.4 AIM ....................................................................................................................... 5 1.4a Specific Objectives .......................................................................................... 5 CHAPTER TWO ............................................................................................................ 6 LITERATURE REVIEW ............................................................................................... 6 2.1 MALARIA ............................................................................................................ 6 2.1.1 Historical notes ............................................................................................... 6 2.1.2 Etiology .......................................................................................................... 6 2.1.3 Life cycle ........................................................................................................ 7 2.1.4 Transmission ........................................................................................... 10 2.1.5 The Vector .............................................................................................. 11 2.1.6 Clinical manifestation ............................................................................. 11 2.1.7 Epidemiology .......................................................................................... 12 2.1.8 Diagnosis ................................................................................................. 13 2.1.9 Treatment ................................................................................................ 14 University of Ghana http://ugspace.ug.edu.gh v 2.2 BACTERAEMIA ........................................................................................... 15 2.2.1 Bacteraemia episodes .............................................................................. 15 2.2.2 Common etiologic agents of bacteraemia ............................................... 16 2.2.3 Diagnosis of bacteremia .......................................................................... 20 2.2.4 Management of bacteremia .................................................................... 20 2.3 MALARIA AND BACTERIAL CO-INFECTION ....................................... 21 CHAPTER THREE ...................................................................................................... 23 METHODOLOGY ....................................................................................................... 23 3.1 STUDY DESIGN ................................................................................................ 23 3.2 STUDY SITES .................................................................................................... 23 3.3 STUDY POPULATION ..................................................................................... 24 3.3.1 Inclusion Criterion ........................................................................................ 24 3.3.2 Exclusion Criteria ......................................................................................... 24 3.4 SAMPLE SIZE DETERMINATION .................................................................. 25 3.5 SAMPLING METHODOLOGY ........................................................................ 25 3.6 CONSENT AND QUESTIONNAIRE ................................................................ 25 3.7 SAMPLE COLLECTION AND PROCESSING ................................................ 25 3.8 LABORATORY ANALYSIS ............................................................................. 26 3.8.1 Haematology ................................................................................................. 26 3.8.2 Parasitological processing ............................................................................ 26 3.8.3 Widal test ...................................................................................................... 27 3.8.4 Blood culture ................................................................................................ 28 3.8.5 Stool culture .................................................................................................. 28 3.8.6 Biochemical identification ............................................................................ 28 3.8.7 Biochemical characterization (API 20E) ...................................................... 29 3.8.8 Antimicrobial Susceptibility Testing (AST) .......................................... 30 3.9 DATA HANDLING AND STATISTICAL ANALYSIS ................................... 31 University of Ghana http://ugspace.ug.edu.gh vi CHAPTER FOUR ......................................................................................................... 32 RESULTS ..................................................................................................................... 32 4.1 ENROLMENT ............................................................................................... 32 4.2 STUDY PARTICIPANTS ............................................................................. 33 4.3 CLINICAL CHARACTERISTICS ................................................................ 34 4.4 LABORATORY FINDINGS ......................................................................... 35 4.5 RISK FACTORS ............................................................................................ 38 CHAPTER FIVE .......................................................................................................... 39 DISCUSSIONS ............................................................................................................. 39 5.1 LIMITATAIONS ........................................................................................... 45 2.5 CONCLUSION .............................................................................................. 45 5.3 RECOMMENDATIONS ............................................................................... 46 REFERENCES ............................................................................................................. 47 APPENDIX A: CONSENT FORM ............................................................................. 69 APPENDIX B: QUESTIONNAIRE ............................................................................ 72 APPENDIX C: MEDIA AND STANDARD SOLUTIONS ........................................ 73 APPENDIX D: STAINING PROCEDURES ............................................................... 78 APPENDIX E: BIOCHEMICAL TESTS ..................................................................... 79 University of Ghana http://ugspace.ug.edu.gh vii LIST OF TABLES Table 3.1 Antibiotic Disc Concentrations used for the isolated organisms Table 4.1 Distribution of participants from the three sites studied. Table 4.2 Demographic characteristics of the participants. Table 4.3 Associations between clinical features and co-infection in participants. Table 4.4 Haematological features of single and co-infections among participants Table 4.5 Isolates from positive blood culture Table 4.6a Antimicrobial susceptibility profiles of the Staphylococcus aureus isolates. Table 4.6b Antimicrobial susceptibility profiles of the Enterobactereacae isolates. Table 4.7 Risk factors for invasive bacterial infections. LIST OF FIGURES Figure 2.1. Overview of Plasmodium's life cycle Figure 2.2. Global distribution of Malaria. SOURCE: WHO, 2012. Figure 3.1 Map of Greater Accra Region showing the 16 districts University of Ghana http://ugspace.ug.edu.gh viii LIST OF ABBREVIATIONS API Analytical Profile Index BA Blood Agar BD Becton and Dickinson BF Blood Film BHI Brain Heart Infusion CA Chocolate Agar CDC Centre for Disease Control EDTA Ethylene Diamine Tetraacetic Acid GHS Ghana Health Service GSS Ghana Statistical Service NTS Non-Typhoidal Salmonella iNTS Invasive Non-Typhoidal Salmonella Mac Mackonkey NMCP National Malaria Control Programme PMI President’s Malaria Initiative PML Princess Marie Louis RDT Rapid Diagnostic Test SS Salmonella-Shigella UNICEF United Nations International Children’s Fund WHO World Health Organization University of Ghana http://ugspace.ug.edu.gh ix ABSTRACT Background: Malaria predisposes children in areas where malaria is endemic to concurrent bacteraemia. In the tropics, co-infections of both diseases are prevalent and are the leading causes of paediatric hospital admissions, morbidity and mortality. Methods: A cross-sectional study was conducted to investigate the prevalence of co- infection of malaria and bacterial bloodstream infections among 232 children under 13 years who reported to three healthcare facilities in Accra and Dodowa with conditions of febrile illnesses suspected to be malaria. The study was conducted between the months of May and December 2014. Results: Out of 1187 eligible febrile children, only 232 (19.55%) who tested positive for malaria were included in the study. They comprised 121 males and 111 females. Blood and stool specimens were taken for haematological analysis and culture for the identification of pathogenic bacteria after malaria diagnosis. Descriptive data were summarised and chi-square analysis was used in testing for associations. Fever (76.72%), anaemia (69.39%) and vomiting (49.56%) were the commonest symptoms of clinical visits. Of the 232 children tested, blood cultures were positive in 5.6% (13/232) for bacterial agents and there were no bacteria isolated from stool cultures. Anaemia, parasitaemia and white blood cell counts were high but not associated with co-infection after chi-square analysis. Co-infection of malaria and bacteraemia was associated with children who never patronised food from outside their homes. Other risk factors were in high frequencies but were not associated with co-infections. Conclusion: These results may suggest co-infection of bacteraemia and malaria, however non-typhoidal Salmonella may not be associated with malaria in the present study. University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTION 1.1 BACKGROUND Febrile illnesses remain the leading cause of paediatric mortality and morbidity especially in sub-Saharan Africa (Bryce et al., 2005; Crill et al., 2006). According to the World Health Organization (WHO), febrile illness is an acute illness characterized by a rise in body temperature. Malaria and bacteraemia are among the commonest causes of these febrile illnesses and are of major public health importance in developing countries (WHO/CDR 1995; Berkley et al., 2005; Bryce et al., 2005; Uneke, 2008). Worldwide, an estimated 76% of the under-fives’ deaths occur due to undiagnosed invasive bacterial infections (Christopher et al., 2013), and in Africa, bloodstream bacterial infections are responsible for 1 out of 6 deaths in children before their fifth birthday (Blomberg et al., 2007). Berkley et al. (2005), have documented that, in malaria endemic areas, 11% of the children admitted with fever are found to have bacteraemia and 12% of these children will die because malaria was over diagnosed at the expense of other causes of fever. Malaria, a mosquito borne infectious disease infects both humans and primates, and is caused by parasites of the genus Plasmodium (Warren, 1993). Globally, it remains the most important disease in tropical and sub-tropical countries, posing a huge burden on health and economic development. It has also been a major obstacle to sustainable development by the world’s poorest regions (Gallup and Sachs, 2001). Approximately 198 million cases of malaria were reported at the end of 2013 with 584,000 deaths (Bassat et al., 2015). University of Ghana http://ugspace.ug.edu.gh 2 Most bacterial infections are widespread but more prevalent in regions where sanitary conditions are poor and may invade the bloodstream after a wide variety of focal infections. Transient bacteraemia is usually non-alarming but may progress to septicaemia which can be life-threatening when immediate medical attention is not given (Meremikwu et al., 2005). Septicaemia is a bloodstream infection usually caused by pathogenic bacteria and together with bacteraemia may be collectively referred to as invasive bacterial infections. Varieties of bacteria found to cause febrile illnesses in children include Staphylococcus spp, Streptococcus spp, Enterobacter spp, Escherichia coli, Klebsiella pneumoniae, Pseudomonas spp, Enterococcus spp, Neisseria meningitides, Salmonella spp, Moraxella catarrhalis, Haemophilus influenzae and Campylobacter spp (Bandyopadhyay et al., 2002; Tintinalli et al., 2004; Wald and Minkowski, 1980). Among the commonly reported bacterial etiologic agents isolated from African children with bacteremia are; Salmonella species, Streptococcus pneumonia and other Gram-negative bacteria (Bronzan et al., 2007; Were et al., 2011; Shaw, 2008; Graham, 2000; Walsh et al., 2000; Enwere et al., 2006; Roca et al., 2006; Sigauque et al., 2009). Several studies have associated invasive bacterial infections with high mortality in children with severe malaria in sub-Saharan Africa (Bronzan et al., 2007; Berkley et al., 2009; Were et al., 2011). Both malaria and bacteraemia mainly affect young children in the sub region and represent the principal cause of hospital admission, hence a massive burden for the under-resourced health facilities. Together, they account for more than half of all paediatric cases on admission to hospitals (Berkley et al., 2005). Comparing all co-infections of malaria and bacterial agents, non-typhoidal Salmonella (NTS), a species of Salmonella is consistently reported as the main bloodstream bacterial infections seen in African children with severe P. falciparum malaria (Gordon et al., University of Ghana http://ugspace.ug.edu.gh 3 2008, Bronzan et al., 2007; Oundo et al., 2002; Mackenzie et al., 2010; Mtove et al., 2011). Malaria has long been alleged to increase the risk of bacterial infections and may contribute especially to the seasonality of NTS disease (Smith, 1982; Kariuki et al., 2006). The clinical presentations of children with bloodstream infection are generally; fever, difficult breathing, tachycardia, malaise, inability to feed or lethargy, although those with asymptomatic bacteraemia are not likely to show any signs of illness (Meremikwu et al., 2005). Since antiquity, clinicians have had difficulty in differentiating invasive bacteraemia from malaria based on clinical presentations alone due to these overlapping clinical features especially in the early stages (Cox et al., 1996; Nsutebu et al 2002). Their differentiation thus requires appropriate laboratory investigations for confirmations as fever or changes in body temperature is frequently associated with malaria in many endemic areas and therefore treated accordingly. This may have led to considerable overestimation of the incidence of malaria whereas bacteraemia remains unsuspected and the causative agent unrecognised (Evans et al., 2004). 1.2 PROBLEM STATEMENT About 20-50% of all hospital admissions are a consequence of malaria with high case- fatality rates due to late presentation and inadequate management (WHO/UNICEF, 2003). Malaria and NTS results in substantial burden of illness and death (Morpeth, 2009). In malaria endemic regions, the presence of bacterial etiologic agents in addition to favourable environmental factors such as improper sewage disposal, poor personal hygiene, poverty, and rapidly increasing urbanization may facilitate co-infection of these diseases (Morpeth, 2009; Keong and Sulaiman, 2006). University of Ghana http://ugspace.ug.edu.gh 4 Ghana records nearly 25-40% of all outpatient clinic visits for malaria, most of which is diagnosed clinically (WHO/UNIFEC, 2005). The presentations of malaria which include fever and general weakness are nonspecific and may well be due to other bacterial infections (Luxemburger et al., 1998). It is difficult clinically to differentiate malaria from bacterial infections such as NTS or typhoid without appropriate laboratory investigations. Moreover, many facilities lack the laboratory capacity to undertake bacterial culture in routine investigations of these bacterial infections. In Africa, including Ghana, most cases of malaria and bacterial co-infections are diagnosed on the basis of clinical symptoms and treatment is presumptive without laboratory confirmation. 1.3 JUSTIFICATION / RATIONALE Predictors of positive blood culture is crucial for clinicians to ensure a timely and appropriate management response. The present study will provide an epidemiological data on co-infection of malaria with bacterial agents in Accra. Data from this study may aid development of preventive strategies including active surveillance systems to control and manage co-infections as well as contribute to implementation of the guided empirical treatment of common bacteria isolates causing febrile illnesses at the study sites. University of Ghana http://ugspace.ug.edu.gh 5 1.4 AIM To investigate the prevalence of malaria and bacterial co-infections in children. 1.4a Specific Objectives  To determine the prevalence of malaria and bacterial bloodstream infections in the study population.  To determine the haematological indicators of single and co-infections of malaria and bacteria bloodstream infections  To determine the risk factors associated with malaria and bacterial co-infections. University of Ghana http://ugspace.ug.edu.gh 6 CHAPTER TWO LITERATURE REVIEW 2.1 MALARIA 2.1.1 Historical notes Malaria derived its name from the Italian word “Mal’aria” which means “bad air”, as the disease was associated with marshy areas. Malaria is an ancient disease and was previously described in Chinese medical writing (CDC, 2012). Some other earlier references to the disease include the accounts of the Hippocrates who described the symptoms of Malaria (Boyd, 1949). In 1880 Charles Louis Alphonse Laveran, a French Army surgeon in Algeria discovered Malaria parasites in the blood of a patient and 18 years later, Dr Ronald Ross, a British medical officer in India discovered that the causative agent of malaria was transmitted by mosquitoes. Subsequently, Giovanni Battista Grassi, an Italian Professor confirmed the vector to be Anopheles mosquitoes (CDC, 2012). 2.1.2 Etiology Malaria is caused by intraerythrocytic protozoan parasites belonging to Plasmodium species (phylum Apicomplexa). Human malaria is caused by five different species of Plasmodium: P. falciparum, P. malariae, P. ovale, P. vivax and P. knowlesi (Alam, 2014). The species differ in their geographical distribution, morphology, immune response, relapse patterns and drug response. P. falciparum causes tropical malaria, P. vivax and P. ovale cause tertian malaria whilst P. malariae causes quartan malaria (Harinasuta et al., 1988). Relapses are characteristic in P. vivax and P. ovale infections. University of Ghana http://ugspace.ug.edu.gh 7 The most widespread species are P. vivax and P. falciparum, the latter is attributable to the severest forms of malaria whilst infections of other species are rarely life-threatening (Carpenter et al., 1991; Breman, 2004). P. ovale is restricted to West Africa sub-region whereas P. malariae is found worldwide at low prevalence (Carter and Mendis, 2002). Occasionally, humans become infected with a zoonotic species, P. knowlesi (Daneshvar et al., 2009; Chin, et al., 1968; Cox-Singh et al., 2008). 2.1.3 Life cycle The malaria parasite has a complex life cycle divided into three stages; the exo-erythr ocytic or pre-erythrocytic stage which usually occurs in the liver, the erythrocytic stage which occurs in the erythrocytes, and the sexual stage (sporogony) which occurs in the mosquito. The exo-erythrocytic and the erythrocytic stages constitute the asexual cycle (schizogony) (Figure 2.1). Figure 2.1 Overview of Plasmodium's life cycle. (CDC, 2006). University of Ghana http://ugspace.ug.edu.gh 8 2.1.3a Schizogony (Asexual stage) 2.1.3a1 Exo-erythrocytic cycle (Human Liver stages) Infective sporozoites from the salivary gland of the female Anopheles mosquito are introduced into the human bloodstream during a blood meal. From the bloodstream, the sporozoites invade hepatocytes where they remain for one or two weeks (prepatent period) and undergo asexual replication known as exo-erythrocytic schizogony to develop into schizonts (NIH, 2007; Khan and Lai 1999). The schizonts contain thousands of merozoites which are released into the bloodstream. It is estimated that each Plasmodium falciparum sporozoite can give rise to up to 40,000 merozoites. In P. vivax and P. ovale, some injected sporozoites may differentiate into stages called hypnozoites which may remain dormant in the liver cells for some time only to undergo schizogony causing relapse of disease when the red cells are invaded. This period of maturation is usually not accompanied by any clinical illness (Khan andLai, 1999). 2.1.3a2 Erythrocytic cycle (Human Blood stages) Merozoites released from schizonts invade erythrocytes in the bloodstream, within 1-2 minutes. According to Miller et al. (2002), merozoites enter erythrocytes by a complex invasion process divided into four phases: (a) initial recognition and reversible attachment of the merozoite to the erythrocyte membrane; (b) reorientation and junction formation between the apical end of the merozoite (irreversible attachment) and the release of substances from the rhoptry and microneme organelles, leading to formation of the parasitophorous vacuole; (c) movement of the junction and invagination of the erythrocyte membrane around the merozoite accompanied by removal of the merozoite's surface coat; and (d) resealing of the parasitophorous vacuole and erythrocyte membranes after completion of merozoite invasion (Miller et al., 2002; Tuteja, 2007). University of Ghana http://ugspace.ug.edu.gh 9 Upon entry, the merozoites use hemoglobin as source of energy and this point are transformed into trophozoites. The early trophozoite is often referred to as the ‘ring form’, because of its characteristic morphology (Figure 2.1). The trophozoite enlarges and is accompanied by highly active metabolism including glycolysis of large amounts of imported glucose, the ingestion of host cytoplasm and the proteolysis of hemoglobin into constituent amino acids. Malaria parasites cannot degrade the heme by-product and free heme is potentially toxic to the parasite. Therefore, during hemoglobin degradation, most of the liberated heme is polymerized into hemozoin. It then undergoes multiple rounds of nuclear division without cytokinesis resulting in the formation of schizonts (Miller et al., 2002). Each mature schizont produces up to about 36 merozoites and these are released after lysis of the RBC to invade other uninfected RBCs (NIH, 2007). The release of erythrocytic merozoites coincides with the sharp increases in body temperature during the progression of the disease, and the repetitive intra-erythrocytic cycle of invasion– multiplication–release–invasion continues, taking about 48h in P. falciparum, P. ovale and P. vivax infections and 72h in P. malariae infection (Tuteja, 2007). This occurs somewhat synchronously and the merozoites are released at approximately the same time of the day and continue until it is brought under control by the immune system or by antimalarial drugs. The contents of the infected RBC that are released upon its lysis stimulate the production of tumor necrosis factor (TNF) and other cytokines, which are responsible for the characteristic clinical manifestations of the disease (Tuteja, 2007). Merozoites of some Plasmodium species show a distinct preference for erythrocytes of certain age. For instance, merozoites of P. vivax attack young immature RBCs called reticulocytes, those of P. malariae attack the older erythrocytes while those of P. falciparum indiscriminately enter into any available erythrocyte (Aikawa et al., 1980). University of Ghana http://ugspace.ug.edu.gh 10 A small proportion of the merozoites in the red blood cells eventually differentiate to produce micro- and macrogametocytes after a variable number of cycles. The gametocytes have no further activity within the human host but are essential for transmitting the infection to new hosts through female Anopheles mosquitoes (Figure 2.1). 2.1.3b Sporogony (sexual stage) - Mosquito stages A mosquito taking a blood meal from an infected individual may ingest gametocytes into its midgut. These gametes fuse and undergo fertilization which occurs in the mosquito’s stomach, producing zygotes. The zygotes develop into motile, elongated ookinetes, which penetrate the mosquito’s mid-gut wall and develop into oocysts. The oocysts grow, divide, and rupture, releasing sporozoites that travel to the mosquito’s salivary glands for onward transmission into another host (Figure 2-1). The sporzoites are found in the salivary glands after 10–18 days and thereafter the mosquito remains infective for 1–2 months. Thus the infectious cycle can repeat once the mosquito feeds on another human host (Tuteja, 2007). 2.1.4 Transmission Malaria is transmitted through; the injection of sporozoites during mosquito bites, blood transfusion and vertical transfer of the parasites from infected mothers to their children before or during birth (congenital malaria) (Hoffman, 1996). However, the main mode of transmission responsible for majority of the cases seen worldwide is through the bite of an infected female Anopheles mosquito. University of Ghana http://ugspace.ug.edu.gh 11 2.1.5 The Vector Out of 400 species of Anopheles mosquitoes, about 60 are capable of transmitting malaria under natural conditions, 30 of which are of major importance. The major vectors are Anopheles gambiae species complex and A. funestus. These species generally bite late in the night, are indoor resting, and are most common in the rural and peri-urban areas. Transmission is proportional to the density of the vector, number of times of bites each day, the survival of the vector after feeding and a human host (reservoir). A. gambiae is the most infective vector, they are tough, long lived, naturally occurring in high densities and bite humans frequently. 2.1.6 Clinical manifestation Malaria presents with symptoms such as fever, headache, muscle pain, vomiting, rapid breathing, coughs and convulsions (Warrell et al., 1990). Symptoms may begin with indefinite malaise, a slow rising fever lasting several days, chills, headache, nausea, and ends with profuse sweating. After a period free of fever, the cycle of chills, fever and sweating is repeated every one to three days (Cheesbrough, 1987). Malaria can cause anaemia which may be severe particularly in young children. Severe malaria can cause black water fever, cerebral malaria, pulmonary oedema (rare but often fatal), and hypoglycaemia which is being increasingly reported in patients with severe malaria, especially children and pregnant women (Waller et al., 1995; Murphy and Breman, 2001). University of Ghana http://ugspace.ug.edu.gh 12 2.1.7 Epidemiology 2.1.7a Global Distribution of Malaria Malaria is the 3rd leading cause of death for children under five years worldwide, after pneumonia and diarrheal diseases (WHO, 2013). Malaria affects a wide number of countries and has a broad distribution in both the subtropics and the tropics. Sub- Saharan Africa remains most heavily burdened, other areas of high endemicity include; India, Brazil and Sri Lanka as shown in figure 2. In Africa and some part of India, the disease occurs both in urban and rural areas and is most prevalent during rainy seasons. An estimated 3.3 billion people are at risk with nearly 90% of all malaria deaths occurring among children in Africa. Out of about 198 million cases of malaria recorded worldwide in 2013, 584 000 died (WHO, 2014). Figure 2.2: Global distribution of Malaria. SOURCE: WHO, 2012. University of Ghana http://ugspace.ug.edu.gh 13 2.1.7b Malaria situation in Ghana In Ghana, malaria is perennial in all parts of the country, with seasonal variations that are more pronounced in the Northern sector. Ghana’s entire population of 24.2 million (2010 Census) is at risk of malaria infection (PMI, 2012), but children under five years of age and pregnant women are at higher risk of severe illness due to lowered immunity. Between 3.1 and 3.5 million cases of clinical malaria are reported in public health facilities each year, of which 900,000 cases are in children under five years (USAID, 2009). The WHO recently estimated total malaria-attributable child deaths at 14,000 per year in Ghana (WHO, 2008). The intensity of malaria transmission ranges from May- October in the Northern part of the country but may be longer in the forest zones. Peak levels of malaria infection in the population may persist for two-three months into the dry season (Ahmed, 1989). 2.1.8 Diagnosis Malaria is diagnosed clinically based on a patient’s signs and symptoms upon physical examination. However, the non-specific nature of symptoms, which overlap with other common infections remain a major challenge to clinical diagnosis. This can impair diagnostic specificity leading to the promotion of indiscriminate use of antimalarial agents (Mwangi et al., 2005; McMorrow et al., 2008; Bhandari et al., 2008b). It is therefore necessary to confirm clinical findings with appropriate laboratory tests. Patients diagnosed with malaria are generally categorized as having either uncomplicated (<250000 parasites/µl) or severe malaria (>250000 parasites/µl). University of Ghana http://ugspace.ug.edu.gh 14 2.1.8a Diagnostic Assays Malaria is diagnosed in the laboratory via microscopy, rapid diagnostic tests (RDT) (Holland et al., 2005) and quantitative buffy coat (QBC) method (Bhandari et al., 2008b). Conventional microscopic diagnosis requires staining thin and thick peripheral blood smears with Giemsa to give the parasites a distinctive appearance. This technique remains the gold standard for laboratory confirmation of malaria. Serological methods such as indirect immunofluorescence (IFA) or enzyme-linked immunosorbent assay (ELISA) do not detect current infection but rather measures past exposure. Molecular diagnostic methods such as polymerase chain reaction (PCR) is most useful for confirming the species of malarial parasite after the diagnosis but are not used routinely due to cost and technicality. 2.1.9 Treatment Uncomplicated malaria may be treated with oral antimalarial agents; severe malaria requires parenteral therapy. The WHO recommends artemisinin combination therapy (ACT) as first line therapy for uncomplicated malaria (USAID, 2009) and parenteral artesunate (a derivative of arthemisinin) is the most potent agent for the treatment of severe malaria (Dondorp et al., 2005a; Dondorp et al., 2010). Artemisinin is a sesquiterpene lactone extracted from the leaves of Artemisia annua (sweet wormwood), which has been used for centuries in China for the treatment of fever. It is an effective and rapidly acting agent for elimination of blood stage parasites, with a broad spectrum of activity against asexual forms from young rings to mature schizonts, as well as gametocytes of P. falciparum. The drug seems to inhibit an essential calcium adenosine University of Ghana http://ugspace.ug.edu.gh 15 triphosphatase, PfATP6, outside the food vacuole of the parasite (Eckstein-Ludwig et al., 2003). 2.2 BACTERAEMIA Bacteraemia is the invasion and circulation of bacteria through the vascular system. A more severe form, septicaemia results when circulating bacteria multiply at a rate that exceeds their removal by phagocytes (Berger, 1983). It is characterized by fever, chills, malaise and toxicity (Parrillo, 1993). Bacteraemia may progress to other infections such as meningitis and endocarditis (Parrillo, 1993) if left untreated. Earlier study in Ghana found that, invasive bacterial infections were associated with a mortality of about 40%, with NTS and Staphylococcus aureus being among the most common organisms isolated (Evans et al., 2004). Streptococcus pneumoniae and Haemophilus influenzae are also responsible for deaths in children with bacterial bloodstream infections (Berkley et al., 2005). These two organisms have also been associated with occult bacteraemia (unsuspected bacteraemia) in apparently healthy children younger than 2 years of age with positive blood cultures (Berger, 1983). 2.2.1 Bacteraemia episodes Bacteraemia may be described as transient, intermittent or continuous depending on their entry into the circulatory system (Mahon et al., 2000). Transient bacteraemia occurs when normal flora are displaced from their usual sites into the blood (LeFrock et al., 1973). Intermittent bacteraemia involves the periodic passing of bacteria from an infected part of the body into the blood. Continuous bacteraemia generally occurs through intravascular infections such as endocarditis or through catheterization and indwelling cannulas (Musher et al., 2000). University of Ghana http://ugspace.ug.edu.gh 16 2.2.2 Common etiologic agents of bacteraemia A wide range of bacteria are responsible for bloodstream infections. These organisms however differ from one locality to the other with varying antimicrobial susceptibility patterns (Meremikwu et al., 2005). Contrary to some studies that have named Gram positive organisms as the commonest isolates in neonates (Phiri et al., 2005), Ayoola et al have established that Gram negative organisms are more common than the Gram positive organisms (Ayoola et al., 2003). In Ghana, NTS and S. aureus are the predominant aetiological agents of bloodstream infections (Nielsen et al., 2012). Some bacteria responsible for bloodstream infections in children are as follows: 2.2.2a Non typhoidal Salmonellae (NTS) Non typhoidal Salmonellae is a common cause of bloodstream infection among African children according to several studies conducted in Africa (Graham et al., 2000; Ikumapayi et al., 2007). The important strains include Salmonella Enteritidis, Salmonella Choleraesuis, and Salmonella Typhimurium (Brooks et al., 2007). NTS is one of the three major causes of invasive disease in children below the age of three (Ikumapayi et al., 2007) resulting in high morbidity and eventual death (Oundo et al., 2002; Vaagland et al., 2004). NTS infections occur worldwide and their mode of transmission is oro-faecal. NTS causes self-limiting gastroenteritis in healthy individuals in developed countries (Kariuki et al., 2006) whilst in sub- Saharan Africa, it causes bloodstream infection in children and adults and may lead to death if prompt and appropriate antimicrobial therapy is not given (Graham et al., 2000; Gordon et al., 2002). A study conducted by Feasey et al. (2012) suggests that fatality from NTS infections ranged from 20 to 25%. Kariuki et al (2006) found out that NTS was responsible for 51.2% of bloodstream infection in Kenya. In Ghana, Nielson et al (2012) and Evans et al., (2004) reported 53.3% and 43% respectively. University of Ghana http://ugspace.ug.edu.gh 17 2.2.2b Salmonella Enterica serova Typhi (Salmonella Typhi) Typhoid fever also known as enteric fever is found worldwide and accounts for several cases of morbidities and mortalities. They are common in developing countries where sanitary conditions are very poor (Evans et al., 2004). Typhoid infections may be mild or severe but can sometimes be life threatening if proper attention is not given. Transmission of enteric fever is oro-faecal through the ingestion of contaminated food or water. In Ghana, typhoid fever is predominant in areas with poor sanitary conditions (Nielsen et al., 2012; Acquah et al., 2013). 2.2.2c Other Enterobacteriaceae Enterobacteriaceae is the general group of bacteria that colonize the gastrointestinal tract. They are the most significant contributors to intestinal infections, which are among the most frequent diseases in the developing world. Examples include Escherichia coli, Klebsiella spp, Enterobacter spp, Salmonella and Shigella (Kayser et al., 2005). Enterobacteriaceae have important pathogenicity factors namely; endotoxins, exotoxins, invasins and colonizing factors which aid in their adaptation in the gastrointestinal tract. When a host is immunosuppressed, Escherichia coli may reach into blood stream and cause sepsis (Mahon et al., 2000). E. coli is among the leading causes of meningitis in infants (Brooks et al., 2007). Klebsiella pneumoniae causes a small proportion (about 1%) of bacterial pneumonias but may also occasionally result in urinary tract infections, bacteraemia and other extra- pulmonary infections. University of Ghana http://ugspace.ug.edu.gh 18 2.2.2d Staphylococcus aureus Staphylococcus aureus is a major pathogen causing pyogenic and toxin mediated infections in humans. It is a part of the normal microbiota of the nose, skin, mouth, and other parts of the body. It is ubiquitous and a common cause of most superficial and invasive infections notably, skin and soft-tissue infections, endovascular infections, septicaemia, endocarditis and wound infections (Del Rio et al., 2009; Hakeem et al., 2013). Bloodstream infections caused by S. aureus are often difficult to treat, and therefore associated with relatively high morbidity and mortality (Shinefield et al., 2002; Naber, 2009). Findings from a study conducted in Ghana by Evans et al. (2004) have reported S. aureus (29%) as one of the major etiologic agents. This agrees with studies conducted by Meremikwu (2005) and Awoniyi et al., (2009) in Nigeria that have also reported S. aureus as a major organism isolated representing 48.7% and 28% respectively. In Mozambique S. aureus (39%) was found as one of the major pathogens isolated from neonates with bloodstream infections followed by group B Streptococcus (20%) (Sigaúque et al., 2009). Another study conducted by Tsering et al (2011) in India also revealed that 97% cases of septicaemia in children was caused by S. aureus. 2.2.2e Streptococcus pneumoniae Streptococcus pneumoniae is an important pathogen that causes diseases ranging from upper respiratory tract infections to severe invasive diseases such as pneumonia, septicaemia and meningitis (Bogaert et al., 2004; Donkor et al., 2013). Transmission of pneumococcus is usually by direct contact with contaminated respiratory secretions and is highest in young children. University of Ghana http://ugspace.ug.edu.gh 19 Streptococcus pneumoniae is the leading cause of death in children less than five years (Isaacman et al., 2010) with about 1.2 million new cases of pneumococci infections emerging annually (Bogaert et al., 2004; Donkor et al., 2013). Nielsen et al. in 2012 conducted a study in Ghana and concluded that S. pneumoniae (9.1%) was among the frequent isolates in blood amongst Ghanaian children. 2.2.2f Haemophilus influenzae H. influenzae belongs to normal bacterial flora of the respiratory tract and a major cause of several invasive and non -invasive infections (García-Cobos et al., 2008). Diseases caused by H. influenzae include childhood pneumonia, meningitis, septicaemia, acute otitis media and epiglottitis (Tristram et al., 2007; Resman et al., 2011). Hib is commonly found in the nose and throat of healthy individuals living in areas where vaccination is not carried out. Almost all children who are not vaccinated are exposed to Hib by the age five. 2.2.2g Neisseria meningitidis Neisseria meningitidis, commonly called meningococcus is found in the mucosa of the oropharynx of humans and a natural colonizer of the upper respiratory tract (Stephens et al., 2007; Caugant and Maiden, 2009). Humans are the only natural reservoir and therefore infection is spread from man to man, the nasopharynx is the site from which meningococci are transmitted through droplet secretions or via aerosols from an infected person to a susceptible individual (Rosenstein et al., 2001). Meningococcal diseases have repeatedly caused epidemics (Greenwood, 2007) and are still a global health problem affecting all ages, and a leading cause of bacterial meningitis and septicaemia (Thompson et al., 2006; Antignac et al., 2003). University of Ghana http://ugspace.ug.edu.gh 20 2.2.3 Diagnosis of bacteremia Prompt diagnosis and effective treatment is often required for bloodstream infection to prevent complications (Meremikwu et al., 2005; Prabhu et al., 2010). Bacteriological examination is therefore very essential for diagnosis of bacteraemia. Although blood cultures remain the gold standard for diagnosis, its’ limitation is that most diagnostic procedures may take up to a week to complete and may cause some clinicians to depend on empirical treatment (Omoregie et al., 2009). 2.2.3a Blood Cultures Blood cultures are mainly employed in the diagnosis of bacteraemia (Prabhu et al., 2010). Physical signs and symptoms may be useful in identifying patients but have limited specificity (Kamga et al., 2011). Culture and isolation of specific pathogen in bacteriological cultures is definitive diagnosis for suspected cases of bacteraemia (Meremikwu et al., 2005). Positive blood cultures, though the gold standard for diagnosing bacteraemia, may give false negative results in neonates even when there are strong clinical suggestions of infection. This is attributable to the fact that antibiotics administered to mothers during pregnancy may suppress the growth of bacteria in culture, yet the neonate may have clinical symptoms and laboratory findings may not indicate bacteraemia or septicaemia (Kaufman and Fairchild, 2004). 2.2.4 Management of bacteremia Antibiotics are used to treat bacterial bloodstream infections worldwide. A wide range of antimicrobials including cephalosporins, aminoglycosides, flouroquinolones and cabarpenems have been successfully used in treatment. Results of bacteriological cultures and antimicrobial susceptibility tests may take about five days, necessitating University of Ghana http://ugspace.ug.edu.gh 21 initial empirical treatment of the infection. This practice however can add up to the already existing problems of antimicrobial resistance. Antimicrobial resistance may occur when selected antimicrobial agents are over prescribed. In order to develop effective guidelines for empirical antimicrobial treatment, knowledge of the type of etiologic agents and the pattern of antibiotic resistance are fundamental (Berkley, 2005). 2.3 MALARIA AND BACTERIAL CO-INFECTION An association between malaria and susceptibility to invasive bacterial infection has been known for almost a century (Dondorp et al., 2005a), and has been repeatedly documented in different settings across Sub-Saharan Africa (Cook and Zumla, 2009; Medana et al. 2011). This association was first described for malaria and non-Typhoid Salmonella (NTS) bacteraemia (Dondorp et al., 2005b), which remains the most frequent cause of malaria associated bacteraemia in many studies, but also includes susceptibility to other Gram negative bacteria. (Cook et al., 2009; Haldar et al., 2007). The linking of NTS to malaria has been documented from studies in Africa (Berkley et al., 1999; Oundo et al., 2002). Oundo et al. (2002) realized in their study in Kenya that, septicaemia infection caused by some species of Salmonella were mostly common and severe at peak season of malaria than any other time. It has been observed that malaria infection was often associated with NTS bacteraemia even in countries where NTS infection was very rare in healthy individuals (Brown et al. 2001). Supporting the concept that the malaria was the cause of the susceptibility to NTS infection, observations in British Guyana demonstrated that once malaria was cured with quinine, co-infected individuals were often able to spontaneously clear NTS infection without additional treatment. University of Ghana http://ugspace.ug.edu.gh 22 Studies of the epidemiology of malaria-NTS co-infection have clearly shown that the incidence of malaria and NTS bacteraemia are strongly correlated (Cook et al., 2009; Medana et al., 2011; Maneerat et al., 2000) whereas stool carriage of NTS is not as closely related to the incidence of NTS bacteraemia. Where malaria transmission has declined over time, similar trends have been observed in NTS bacteremia (Cook et al., 2009; Maneerat et al., 2000). In Kenyan children, nearly two-thirds of cases of bacteraemia were attributable to the effect of malaria when malaria transmission was at its highest levels (Cook et al., 2009). NTS has been reported as one of the most common causes of community acquired bacteremia in children presenting to hospital in Kenya, second only to S. pneumoniae. However, the association between malaria and bacteremia extends only to NTS and some other common Gram negative organisms. Gram positive bacteria (Cook et al., 2009) have not been implicated. High case fatality rates have been reported for patients hospitalised with malaria and bacterial co-infections, suggesting that mortality may be increased, (Haldar et al., 2007; Medana et al., 2002; Garcia et al., 1999) Clinical observations have prompted speculation that malaria may cause susceptibility to bacteraemia through immunoparesis (Haldar et al., 2007) impairment of phagocytic cell function (Maude et al., 2009; Essuman et al., 2010) complement consumption (Maude et al., 2009) or increased gut permeability (Haldar et al., 2007). Several subsequent studies have suggested that increased susceptibility to NTS bacteraemia may persist after clearance of microscopically detectable malaria infection (Brown et al., 2001; Haldar et al., 2007; White et al., 2009) or that susceptibility is greater at moderate than high parasite density (White et al., 2009; Maude et al., 2009). Other studies have suggested that the association is particularly strong in the case of severe malarial anemia (Medana et al. 2011; Dorovini et al., 2011; Garcia et al, 1999; Maude et al., 2009). University of Ghana http://ugspace.ug.edu.gh 23 CHAPTER THREE METHODOLOGY 3.1 STUDY DESIGN The study was a cross-sectional investigation conducted between the months of May and December 2014 among children who reported to various healthcare facilities in and around Accra with conditions of febrile illnesses and were suspected to have malaria. Prior to commencement of the study, ethical clearance was sought from the College of Health Sciences, University of Ghana and the Ghana Health Service. 3.2 STUDY SITES The study was conducted in three hospitals within the Greater Accra region, Ghana. The region is divided into 16 districts with a population of 4.01 million. Two of the study sites, the Maamobi and the Princess Marie Louis Children’s (PML) Hospitals are located within the Accra metropolitan area whilst the Shai Osu-Doku District Hospital in Dodowa is located in the Tema metropolis. The sites combined have a total bed capacity of about 300 and a doctor to patient ratio of 1:5,000 (Ministry of Health, Ghana - unpublished reports). All the three facilities provide healthcare services to in-patients and out-patients under various specialties including medicine and surgery. The Princess Marie Louis Children’s (PML) Hospital serves as one of the main paediatric referral centres in the country. It sees approximately 200 out-patient cases daily in all specialties. University of Ghana http://ugspace.ug.edu.gh 24 Figure 3.1: Map of Greater Accra Region showing the 16 districts (Source: Dodowa Malaria Research Center). 3.3 STUDY POPULATION The population investigated were febrile children below 13 years of age seeking medical care at both inpatient and outpatient departments. The participants were selected according to the following criteria: 3.3.1 Inclusion Criterion Children below 13 years presenting with malaria or symptoms of fever whose parents consented. 3.3.2 Exclusion Criteria The study excluded children above 13 years, children on antibiotics and anti-malarial therapy, children of parents who did not consent, and those in critical conditions as severe anaemia. University of Ghana http://ugspace.ug.edu.gh 25 3.4 SAMPLE SIZE DETERMINATION Sample size was calculated using n = z²p (1-p) / e² Where n = the minimum sample size, z = the standard score, p = the known prevalence of co-infection in children and e = the allowable error margin. Using the prevalence of 11% (Berkeley et al, 2005) at 95% Confidence level (z = 1.96, e = 5%), the minimum number of study participants that were enrolled for the study was 100. 3.5 SAMPLING METHODOLOGY Health personnel at the three study sites assisted in educating parents of all eligible participants and addressed questions and concerns. Participating children were then stratified into malarial and non-malarial cases after screening with RDT. All participants who have sufficient data for inclusion criteria were recruited into the study. 3.6 CONSENT AND QUESTIONNAIRE Participation in the study was on a voluntary basis. Written informed consent was obtained from parents and guardians who were willing to include their wards. The children with their guardian were assured of confidentiality of the information they provided. A structured assessment form was used to obtain the clinical history regarding febrile illness, demographic data and risk factors (Appendix A and B). 3.7 SAMPLE COLLECTION AND PROCESSING Three (3) mL of venous blood was collected from each patient aseptically with a needle and syringe. Out of this, 1mL was transferred into a sterile EDTA tube for University of Ghana http://ugspace.ug.edu.gh 26 haematological analysis, malaria diagnosis and serological analysis (widal test). The remaining 2mL of blood was directly inoculated into 18 mL of Thioglycollate broth (Oxoid, UK) for blood culture. Fresh stool samples were collected from the patients into Selenite F broth (10 mL/tube), and the tubes containing samples were transported to the laboratory for bacteriological culture. 3.8 LABORATORY ANALYSIS 3.8.1 Haematology The haematological parameters for each patient were measured using automated haematology analyzer (sysmex 21N, Germany). These included haemoglobin level, total white blood cell (WBC) counts, total red blood cells (RBCs) counts, mean cell volume (MVC), platelet counts, neutrophil counts, lymphocyte counts and eosinophil counts. 3.8.2 Parasitological processing 3.8.2a Rapid diagnostic test The First Response (Premier Medical Corp., India) rapid diagnostic test (RDT) kit was used in screening all blood samples for malaria. In performing this test, 20µL of whole blood and 10µL of a buffer were added to the sample well of the test kit. It was incubated for 15 minutes and the results were read immediately. Positive tests had a purple band on both the test and control regions. The negatives had no bands on the test regions. University of Ghana http://ugspace.ug.edu.gh 27 3.8.2b Thick and thin blood smears Thick and thin films of peripheral blood were prepared to examine malaria parasites. For the thick film, a small drop (5-8uL) of blood was placed at one end of a microscope slide, evenly smeared and dried to identify malaria parasites. Thin blood films were prepared similarly, however, the small drop of blood at one end of the microscope slide was evenly spread out to cover almost the entire length of the slide using a spreader. The blood films were thoroughly air-dried and the thin film was fixed with absolute methanol for species identification (Appendix D). The blood films were then stained with freshly prepared 20% Giemsa solution (in phosphate buffer) left to stain for 20 minutes, followed by rinsing carefully under slow running tap water. The slides were air-dried and examined with immersion oil under light microscope (X100 objective –Primo-Star Zeiss, Germany) for the presence of malaria parasites and species identification. Malaria parasites were counted against 200 White Blood Cells (WBCs) to obtain the parasite density expressed as parasite per microliters (µL) of blood. 3.8.3 Widal test The widal agglutination test was performed on all blood samples by the rapid slide titration method (Lynch and Raphael, 1983) using commercial antigen suspension (Cypress Diagnostic, Belgium) for the somatic (O) and flagella (H) antigens. To perform this test, 50µL of test serum was placed in 2 circles on a glass slide and equal volumes each of positive controls and normal saline also added in different circles. A drop each of O or H antigens were added to the test serum in each circle and then to the negative and positive controls. The content of each circle was mixed and spread to the University of Ghana http://ugspace.ug.edu.gh 28 entire circle after which it was rocked gently for 1 minute and observed for agglutination. 3.8.4 Blood culture Blood culture was done manually by inoculation into Thioglycollate broth and incubated aerobically at 35°C for 7 days and examined visually daily for evidence of bacterial growth. Indicators of bacterial growth that were used included; turbidity of blood-broth mixture, growth of microcolonies, haemolysis, colour changes and gas production. After 24hours of incubation, all cultures showing growth or no growth were sub- cultured onto solid media plates of MacConkey agar, and Blood agar and incubated at 37°C aerobically for 24 hours. In the case where Thioglycollate broth showed no growth up to day 7, subcultures were repeated from the broth on day 7 before it was discarded. All isolates from the subcultures were Gram stained and identified. 3.8.5 Stool culture Stool samples were cultured on Salmonella-Shigella (SS) agar after pre-enrichment in Selenite F broth as described by Brooks et al. (2006). The cultures were incubated for 24 hours at 37ºC and observed for the growth of non-lactose fermenters. 3.8.6 Biochemical identification Colonies from solid agar plates were subjected to biochemical tests and further identification by the API 20E (bio-Mérieux, Inc. France). TSI agar was used to determine the ability of bacteria to ferment glucose and/or lactose and their ability to produce hydrogen sulphide or other gases. Presumptive colonies have alkaline (red) slants and acid (yellow) butts, with or without H2S production (blackened agar). For University of Ghana http://ugspace.ug.edu.gh 29 urease test, the production of ammonia from urea was shown by a change in the phenol red indicator from yellow to pink. Salmonella species are typically urease-negative. Oxidase test was used to determine the presence of an enzyme cytochrome oxidase, which catalyses the oxidation of reduced cytochrome by molecular oxygen. In Simmon’s citrate agar, the use of citrate as a sole carbon source was indicated by the production of ammonia and a change in the colour of the medium from green to blue. Testing for indole production is important in the identification of Enterobacteria. Indole production was tested by Kovac’s reagent and a positive test was indicated by red coloured compound. Motility was indicated by turbidity extending out from the line of stab inoculation. Non-motile organisms grew only in the inoculated area. Gram positive bacteria were identified using the coagulase and catalase test. Catalase activities were detected with BD catalase reagent droppers (BD, Maryland, USA), according to the manufacturer’s instruction (Appendix D). 3.8.7 Biochemical characterization (API 20E) Further confirmation of the isolates was carried out with a commercial bacterial identification kit such as the Analytical Profile Index, API 20E strip kit (bio-Mérieux, Inc., France). The API is a miniaturized panel version of the conventional procedures used for the identification of Enterobacteria and other Gram-negative bacteria. The reagents used included API NaCl 0.85% medium, API 20E reagent kit, Zn reagent, oxidase and mineral oil. To prepare the strips, an incubation box (tray and lid) was used and 5 ml of distilled water were distributed into microcupules of the tray to create humid atmosphere. To prepare the inoculums, single well isolated colony was removed from an isolation plate and emulsified in 5 ml of AI 0.85% NaCl in order to achieve a homogeneous bacterial suspension. Anaerobiosis in the tests arginine dihydrolase University of Ghana http://ugspace.ug.edu.gh 30 (ADH), lysine decarboxylase (LDC), ornithine decarboxylase (ODC), H2S and urea(URE) was maintained by overlaying with mineral oil. The incubation box was closed and incubated at 36±2°C for 24 h as described by the manufacturer. The strains were identified and the number codes generated were interpreted on the results obtained with the API 20E kit using the identification chart supplied by the manufacturer. For quality Control, Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853 and methicillin resistant Staphylococcus aureus ATCC 9213 were set up together with the test organism to control media, biochemical tests, and potency of antibiotic discs. 3.8.8 Antimicrobial Susceptibility Testing (AST) Susceptibilities to various antimicrobial testing was carried out on Mueller Hinton agar as described by the Kirby–Bauer disc diffusion method (Bauer et al., 1966) and interpreted by the Clinical Laboratory Standards Institute guidelines (CLSI, 2013). 3.8.8a Inoculum preparation for AST Plates and antibiotic discs were brought to room temperature before use. Four to five colonies were touched with a straight wire loop and emulsified in peptone until the turbidity was similar to that of 0.5% McFarland standard. 3.8.8b Inoculation and Application of Antibiotic discs A sterile cotton swab was dipped into the inoculum and rotated against the wall of the tube to remove excess volume of the inoculum. The entire surface of the agar plate was swabbed evenly in three directions. The inoculated plates were air-dried, and antibiotic discs (Oxoid, UK) were placed on the agar using flamed forceps and were gently pressed University of Ghana http://ugspace.ug.edu.gh 31 down to ensure contact. Once applied, the discs were not removed. The antibiotic discs used are as shown in table 3.1. 3.8.8c Incubation and Reading Plates were incubated aerobically at 37°C and diameters of zones of inhibitions were measured with a caliper after the 24-hour incubation period. Measured zones of inhibitions were compared with zone diameter interpretative chart (CLSI, 2013). Table 3.1 Antibiotic Disc Concentrations Used for the isolated organisms Gram Positive Bacteria Gram Negative Bacteria Antibiotic Concentration (µg/disc) Antibiotic Concentration (µg/disc) Cefoxitin 30 Ampicillin 25 Penicillin 10 Tetracycline 50 Erythromycin 15 Cotrimoxazole 25 Cefuroxime 30 Gentamicin 10 Gentamicin 10 Cefuroxime 30 Ciprofloxacin 5 Chloramphenicol 30 Ceftriaxone 30 Meropenem 10 3.9 DATA HANDLING AND STATISTICAL ANALYSIS All data were handled confidentially. Clinical and laboratory data sheets were completed by the investigator only. Data was stored in bound folders and put under lock until data entry. During data entry, database files were protected under password. All data was cross-checked for correction of errors that might arise during the course of data entry. Data was entered into a database and analysed descriptively using MS excel and MS access. Measures of central tendency, frequency tables and bar charts were used in data summary. University of Ghana http://ugspace.ug.edu.gh 32 CHAPTER FOUR RESULTS 4.1 ENROLMENT In total, 1187 children presenting with fever, convulsion, anaemia, diarrhoea and vomiting reported to health centres at the three sites during the span of the study. An initial 246 children were recruited but excluding incomplete questionnaires and inadequate specimens provided, 232 valid participants were enrolled. All enrolled participants were malaria positive cases diagnosed through preliminary RDT screening and then confirmed through microscopy. Consenting parents or guardians completed questionnaires while participants provided blood and stool for cultures. The participation for the study at the three sites ranged from 15.52% (Maamobi) to 58.19% (Dodowa) as shown in table 4.1. One hundred and ninety-seven participants (84.91%) were from urban communities whilst 35(15.09%) came from peri-urban communities few kilometres from the study sites. Table 4.1: Distribution of participants from the three sites studied. SITE Number consented Percentage of study participants Dodowa 135 58.19% P.M.L 61 26.29% Maamobi 36 15.52% University of Ghana http://ugspace.ug.edu.gh 33 4.2 STUDY PARTICIPANTS Of the participants, 52.2% were males. The most occurring age group for males was 9- 13 years whilst that of females was 0- 2 years. All participants of school going age (3-5 years for pre-school and 6-13years for basic school) were in school and came from various locations within the environs of the three study sites. The demographic characteristics of the participating children are shown in table 4.2. Table 4.2 Demographic characteristics of the participants. SITE Number Percentage (%) Sex Male 121 52.16 Female 111 47.84 Age group 0-2 60 25.86 3-5 53 22.84 6-8 54 23.28 9-13 65 28.02 Education Schooling 172 74.14 Not schooling 60 25.86 University of Ghana http://ugspace.ug.edu.gh 34 4.3 CLINICAL CHARACTERISTICS The most common malaria associated symptoms leading to hospital visit among the study participants were fever, anaemia and vomiting. There were no significant differences between the clinical presentation of patients and single or co-infection as shown in table 4.3 Table 4.3: Associations between clinical features and co-infection in participants. Co-infections x2 P-value Yes No Severity of parasitemia Mild 2 75 * Moderate 7 93 1.75 0.30 Severe 4 51 1.62 0.23 Fever Yes 11 168 0.46 0.73 No 2 51 * Vomiting Yes 7 108 0.10 0.78 No 6 111 * Anaemia Yes 9 152 0.00 1.00 No 4 67 * Convulsion Yes 0 25 1.66 0.37 No 13 194 * Diarrhoea Yes 5 60 0.75 0.36 No 8 159 * Sickling Yes 0 24 1.42 0.62 No 13 219 * University of Ghana http://ugspace.ug.edu.gh 35 4.4 LABORATORY FINDINGS Of the 232 participants who tested positive for P. falciparum after RDT and microscopy, the median level of parasiteamia was 88,000/µL (range: 6000/µL-380,000/µL). Children aged 9-13 years had higher levels of parasiteamia compared to all other age groups. There were no associations between parasiteamia and clinical presentation. The haematological features of the participants are shown in table 3. Table 4.4: Haematological features of single and co-infections among participants Parasitemia (<250000parasite/µl) Hyperparasitemia (>250000 parasite/µl)) Co-infection (malaria + bacteraemia) n=196 n=36 n=13 Haemoglobin Non-anaemic 35 34 4 Anaemic 145 2 9 Severe anaemia 16 0 0 WBC counts Low 0 0 0 Normal 122 15 5 High 74 21 8 Neutrophil % Low 15 3 2 Normal 133 26 10 High 50 5 1 Lymphocyte % Low 59 22 3 Normal 126 12 8 High 11 2 2 University of Ghana http://ugspace.ug.edu.gh 36 Shown in table 4.4 are the isolates from positive blood cultures. The proportion of blood cultures yielding a clinically significant positive result was 5.6% (13/232) after identification using appropriate biochemical tests and API 20E. Enterobacteria were isolated in 8 positive while the remaining 5 were Staphylococcus aureus. Blood and stool cultures from the participants were negative for salmonella species. All the S. aureus isolates showed complete resistance to penicillin (100%) whiles 80% resistance was for Co-trimoxazole. The Enterobacteriacae were completely resistant to ampicillin. Tetracycline and ceftriaxone also shows high resistance. The etiologic agents of bacteraemia are shown in table 4.4 and their corresponding antimicrobial susceptibility patterns are shown in tables 4.5a and 4.5b. Table 4.5: Isolates from positive blood culture Organisms Total isolates Age groups Gram Negative Bacteria Citrobacter freundii Providencia stuartii Providencia alcalifaciens Enterobacter amnigenus Proteus mirabilis Pseudomonas aucimobilis Enterobacter amnigenus 1 2 1 1 1 1 1 0-2 0-2, 6-8 0-2 3-5 6-8 9-13 9-13 Gram positive Organisms Staphylococcus aureus 5 3-5, 6-8, 9-13 University of Ghana http://ugspace.ug.edu.gh 37 Table 4.6a: Antimicrobial susceptibility profiles of the Staphylococcus aureus isolates. Antibiotics Staphylococcus aureus Cefoxitin 100% S Penicillin 100% R COX 100% S Erythromycin 100% S Co-trimoxazole 80%R Cefuroxime 100%S Gentamicin 100%S Ciprofloxacin 100%S S - Sensitive; R - Resistance. Table 4.6b: Antimicrobial susceptibility profiles of the Enterobactereacae isolates. P stuartii P stuartii P alcalifaciens P aucimobilis E amnigenus E amnigenus P mirabilis C. freundii Ampicillin R R R R R R R R 100%R Tetracycline S R R R R R R R 88%R Cotrimoxazole S R R S S S R S 63%S Gentamicin S S S R R R R S 50%S Cefuroxime S R S S R R S R 50%S Chloramphenicol S S S - S S R S 88%S Ceftriaxone S R R - S S R S 57%S Cefotaxine S R S R R R R R 85%R Ciprofloxacin S S S S S S R R 85%S Amoxicillin S S S S R S S S 88%S Meropenem R S S S S S S R 85%S University of Ghana http://ugspace.ug.edu.gh 38 4.5 RISK FACTORS Table 4.6 shows differences in proportions of the risk factors usually associated with invasive bacterial infections. More than half of the children ate outside their homes frequently, and of those who had toilet facilities at home, only 24.8% (31 out of 125) had flush toilets (water closet). Recent antibiotic use within four weeks prior to the study was also minimal (14.65%). Table 4.7. Risk factors for invasive bacterial infections. *reference Co-infections x2 P-value Yes No Recent antibiotic use Yes 1 33 * No 12 186 0.53 0.70 Sickling Yes 0 24 1.59 0.37 No 13 195 * Toilet facility at home Yes 7 6 0.00 1.00 No 118 101 * Type of toilet facility Bush/polytene 1 14 0.00 1.00 KVIP 10 175 0.37 0.69 WC 2 30 * Eating out Never 7 42 10.28 <0.05 Occasionally 3 44 1.90 0.18 Regularly 3 133 * Water source Packaged 9 105 0.60 1.00 Borehole/tap 4 107 0.26 1.00 Well/Rain 0 7 * University of Ghana http://ugspace.ug.edu.gh 39 CHAPTER FIVE DISCUSSIONS The purpose of this study was to determine the prevalence of co-infection of malaria and bacteraemia and also to assess the risk factors for bacterial infections within selected communities in the Accra sub-metro. Malaria coupled with bacterial infections are a major public health concern as the tendency of misdiagnosis is eminent and therefore tends to increase morbidity and mortality. More children reported to the clinics with febrile illnesses suspected to be malaria than adults. They lived within the environs of the study sites and those who were of school going age were enrolled in various schools. The finding of more children especially from Dodowa and Maamobi adds to anecdotal reports indicating that children are more at risk of febrile illnesses suspected to be malaria compared to adults. Age is an influential factor in malaria infection and may contrast Svenson et al. (1995) who found that the severity of malaria was not age dependent. Health survey from the PML Children’s hospital also indicate that the causes of frequent visits to clinics was febrile related illnesses all of which was suspected malaria. Their increased risk of malaria could be attributable to; immature immunity especially in children under the age of 2 years (WHO, 1996; Rijkers et al., 1998; Klein Klouwenberg and Blont, 2008) and multiple exposures to the vector bites since children may not take necessary precautions about the vector when they engage in outdoors activities. University of Ghana http://ugspace.ug.edu.gh 40 Both diseases have similar symptoms characterized by fever, weakness, body weight loss, anaemia and in some cases gastrointestinal disturbances (Niikura et al., 2008; Samal and Sahu, 1991). The similarity in symptoms is one of the causes of difficulties that arises in the preliminary diagnosis of the diseases and there is no report from the Ministry of Health, Ghana, on the status of co-infections in the country. Bacterial co-infections in malaria positive children was evaluated based on culture and identification with API20E. Thirteen (13) out of the 232 participants had dual malaria and bloodstream bacterial infections, representing 5.6% co-infection. Surprisingly, none of the cultured samples were positive for NTS despite being an important co-infection with malaria and where many of the participants did not have potable water and toilet facilities in their homes. The finding of no positive cultures for NTS is in huge contrast to other studies in the country and elsewhere in Africa that have associated the susceptibility of NTS to malaria infection, while others have related the two as most common in the tropics and subject to misdiagnosis (Evans et al., 2004; Nielson et al., 2012; Bronzan et al., 2007; Lepage et al., 1987; Maitland et al., 2006; Nesbitt et al., 1989). Staphylococcus aureus was the only Gram positive organism among the six different organisms isolated from blood cultures, all others were Gram negative. S. aureus may not necessarily be attributed to malaria, instead, it confirms its implications in community-acquired bacteraemia among children in rural sub-Saharan Africa. Recent investigations elsewhere in Africa show that S. aureus was the most frequent cause of bacteraemia caused by Gram-positive organisms in infants and young children presenting at a hospital in Nigeria (Johnson et al. 2008) and Mozambique (Sigauque et al. 2009). These infections were single-infections and not co-infection with malaria. It also agrees to findings from other developing countries that suggest S. aureus to University of Ghana http://ugspace.ug.edu.gh 41 be a major aetiological agent of septicaemia (Evans et al., 2004; Meremikwu et al., 2005; Hill et al., 2007; Kizito et al., 2007; Komolafe and Adegoke, 2008) which constitute a significant threat to child survival in these areas. The reason why S. aureus was predominant cannot be explained from this study, however, it’s implications in community-acquired bacteraemia among children at a rural sub-Saharan are highlighted. All of the S. aureus were resistant to penicillin (Table 4.5a). Usage of penicillin is therefore not appropriate for the treatment of Staphylococcal systemic infections as it may lead to treatment failure. Cefoxitin, Erythromycin, Co-trimoxazole, Gentamicin and Ciprofloxacin are more effective drugs of choice as the organisms showed no resistance to these groups of antibiotics. Several published reports have also demonstrated that Gram-negative organisms either exceed or rival Gram-positive organisms in bloodstream infections in both adults and children from African countries (Gordon et al., 2002; Ayoola et al., 2003; Archibald et al., 2003) as observed in this study. It is unknown why Gram negative organisms were mostly implicated in bloodstream infections but a likely reason that could explain this occurrence is that the Gram Negative isolates belong to the group of Enterobacteriacae that usually colonize the gut and progress to cause systemic invasion. It is noteworthy that, all the laboratory requisition forms of the participants specified for malaria tests and there were no requests for blood associated bacterial infections. Nevertheless, the isolation of some bacteria in the blood may indicate missed diagnosis which would consequently lead to mistreatment. This may add to growing evidence that, much attention is not paid in cases of malaria related co-infections in Ghana as their clinical symptoms may present in the same way. Systemic bacteraemia reported herein in children with University of Ghana http://ugspace.ug.edu.gh 42 Plasmodium falciparum malaria makes the identification of concurrent infections a prerequisite for adequate treatment given their overlap in clinical signs. With the abundance of affordable antimalarial medicines, self-diagnosis of malaria is often the main diagnostic approach used by many Ghanaians for febrile illnesses. The second clinical diagnosis is often sought after the self-diagnosis and medication has failed (Nyamongo, 2002). While prompt and accurate diagnosis of malaria is part of effective disease management, many other etiologic agents could be responsible for febrile presentations. If these presentations are accurately diagnosed, it could help to reduce indiscriminate use of antimalarial especially in young children and improve differential diagnosis of febrile illness. It is hence important to supplement febrile conditions with blood cultures to ensure accurate diagnosis despite their confounding clinical symptoms. The data presented support accumulating evidence that invasive bacterial infections are important concurrent infections in paediatric populations with malaria. Studies from the Gambia have found community acquired invasive bacteraemia infections to be higher in children aged 2- 29 months compared to adults (Enwere et al., 2006). A similar study at a rural hospital in Mozambique showed an increased risk of community acquired bacteraemia in children <3 years of age compared to adults (Sigauque et al., 2009). These finding of higher bacteraemia in children may not be comparable to results from this study as adults were not included and the few isolates obtained from blood cultures were not enough to draw definitive conclusions. The absence of NTS isolates in positive blood cultures also imply that, results from this study show no significance in associating NTS co-infections with malaria as other authors (Bronzan et al., 2007; Evans et al., 2004) have previously demonstrated. University of Ghana http://ugspace.ug.edu.gh 43 On the other hand, the frequent diagnosis of malaria without routine bacterial cultures to diagnose infections in most primary and secondary health facilities, some clinicians may be tempted to give alternative remedies to other infections that mimic malaria. This may lead to unguided empirical treatment especially with antibiotics, which, in turn, may result into antimicrobial resistance and can affect a patient’s rate of recovery. There is therefore further emphasizes the need for supplementary tests in ensuring accurate diagnosis before commencement of therapy. Anaemia has been shown to be prevalent in areas where malaria is endemic (Le Hung et al., 2005). The association between malaria and anaemia has been well documented (Adam et al., 2005; Huddle et al., 1999; Kagu et al., 2007; Mayor et al., 2007; Muhangi et al., 2007; Ouma et al., 2007; Tarimo, 2007). Malaria is characterized by a drop in the level of haemoglobin, resulting from the destruction of erythrocytes and as expected, the results from this study shows that anaemia was predominant among the participants. In Sub-Saharan Africa where malaria is endemic, the association between anaemia and malaria is so strong that anaemia is often taken as a proxy indicator of the malaria control programmes (Le Hung et al., 2005). Although no mortality was recorded during the span of the study, the prevalence of anaemia was high and may be disadvantageous in malaria positive patients, potentially leading to severe anaemia as more RBCs are invaded and lysed when infection progresses without treatment. It is worth mentioning that the 6.9% severe anaemia observed in of children from this study is a frequent complication of Plasmodium falciparum infections that occurs in young children (Breman, 2001) with a case-fatality rate reaching 23% in malaria holoendemic areas (Obonyo et al., 2007). This study could not establish any associations between malaria and anaemia despite the high proportions of anaemic children. A reason that accounted for this was the lack of a University of Ghana http://ugspace.ug.edu.gh 44 malaria negative control group. However, some studies have reported no association (Le Hung et al., 2005; Stoltzfus et al., 2000; Stoltzfus et al., 1997) between malaria and anaemia. A study by Mato (1998) among Yanomami Amerindian population from the Southern Venezuelan Amazon also found no association between malaria and anaemia. WBC counts are essential in predicting the health status or immunocompetence of a person and are lower to the normal counts during malaria. While increased counts indicate infections, reduced WBC counts, on the other hand, suggest an increased susceptibility to infections. Interestingly, this study found high WBC counts in co-infected children compared to children with only malaria and showed that, the WBC counts of the children did not fall below the standard value as generally anticipated. It contrasts studies that suggest that children who test positive for P. falciparum and those with high parasite densities are associated with low WBC counts (Omalu et al 2008). The reasons for this disparity cannot be discerned from the study, however, a risk factor for NTS infection in malaria is attributable to the reduced WBC counts which impair the defense of the body and subsequently, immunity to diseases. Since no lower counts of WBCs were recorded, it may be a likely reason why NTS was not isolated after culture as it is arguable that the immune system of the children was not impaired to an extent of predisposing them to NTS bacteraemia. This conclusion is however based on thin evidence. Among the risk factors used to access possible transmission of bacterial of NTS among the participants were; parasitaemia, source of drinking water, sickle cell disease, younger age, lack of in-house toilet facilities, and lack of potable water. Surprisingly, participants who never patronized food from outside their homes were associated with bacteraemia (p<0.01). All the other factors showed no association. University of Ghana http://ugspace.ug.edu.gh 45 5.1 LIMITATAIONS This study is subject to several limitations. To begin with, the study was conducted in urban areas which are known to have lesser prevalence of bacterial infections compared to rural centres. Choosing study sites from rural areas would have been preferable for prevalence studies involving NTS. Secondly, molecular techniques involving PCR are more sensitive and would have been more accurate in diagnosis compared to blood cultures since some studies have argued that the sensitivity of culture from 1mL of blood is minimal thus giving only 1CFU/mL for organisms such as NTS. Furthermore, failure to collect other parameters such as temperature, BMI, blood sugar and the comparatively lower sample size are other limitations that prevent the study from drawing a definitive conclusion. 2.5 CONCLUSION In conclusion, malaria has gained more recognition through research compared to bacterial etiologic agents. Similarities in the clinical presentations of both diseases may influence clinicians to pay particular attention to the management of malaria in patients while co-infections may be missed out on the first diagnosis. This study predominantly found S. aureus and Enterobacteria to be responsible for co-infections in malaria patients. NTS which has been previously reported by various authors may not be associated with malaria in the population studied. Although none of the stool cultures was positive for bacteria, there is the need to provide safe drinking water, good toilet facilities and practice adequate personal hygiene in order to University of Ghana http://ugspace.ug.edu.gh 46 prevent the transmission of bacterial infections, especially in areas where malaria is endemic areas. 5.3 RECOMMENDATIONS Knowledge of the prevalence, risk factors of NTS and malaria co-infections in Ghana, particularly in high risk areas is critical in misdiagnosis of malaria regionally and globally. There is the need for clinicians, public health officials GHS/Policy makers to include invasive bacterial infection such as NTS in routine diagnosis of persons presenting with febrile illnesses. Malaria still remains a public health concern especially in children. In the environs of the study sites where drainage is poor and mosquito breeding habitats have an extensive distribution, parents and caregivers should protect their children with long lasting insecticide treated bed nets. Furthermore, preventive strategies among the general public should include; promotion of inbuilt toilet facilities at home to maximise personal hygiene. Provision of potable water for communities without access to running tap water. Traditional microbiological culture methods employed to detect systemic bacteraemia are often time consuming and have modest sensitivity. Therefore, molecular methods may be useful in accurately diagnosing bacteraemia. University of Ghana http://ugspace.ug.edu.gh 47 REFERENCES Acquah, S. E., Quaye, L., Sagoe, K., Ziem, J. B., Bromberger, P. I. and Amponsem, A. A. (2013). Susceptibility of bacterial etiological agents to commonly-used antimicrobial agents in children with sepsis at the Tamale Teaching Hospital. BMC Infect Dis, 13, 89. Adam, I., Khamis, A.H. and Elbashir, M.I. (2005) Prevalence and risk factors for anaemia in pregnant women of eastern Sudan. Transactions of the Royal Society of Tropical Medicineand Hygiene 99, 739‐743 Ahmed, K. (1989). Epidemiology of malaria in Ghana. Ghana Med. J. 23, 190-196. Aikawa, M., Suzuki, M., and Gutierrez, Y. (1980). Pathology of malaria. Malaria, 2, 47-102. Alam, A. (2014). Serine proteases of malaria parasite Plasmodium falciparum: potential as antimalarial drug targets. Interdisciplinary perspectives on infectious diseases, 2014. Antignac, A., Ducos-Galand, M., Guiyoule, A., Pirès, R., Alonso, J. M. and Taha, M.-K. (2003). Neisseria meningitidis strains isolated from invasive infections in France (1999–2002): phenotypes and antibiotic susceptibility patterns. Clin Infect Dis, 37, 912-920. Archibald, L. K., Kazembe, P. N., Nwanyanwu, O., Mwansambo, C., Reller, L. B., & Jarvis, W. R. (2003). Epidemiology of bloodstream infections in a bacille Calmette-Guerin-vaccinated pediatric population in Malawi. Journal of Infectious Diseases, 188(2), 202-208. Awoniyi, D., Udo, S. and Oguntibeju, O. (2009). An epidemiological survey of neonatal sepsis in a hospital in Western Nigeria. Afr J Microbiol Research, 3, 385-389. University of Ghana http://ugspace.ug.edu.gh 48 Ayoola, O. O., Adeyemo, A. A. and Osinisu, K. (2003). Aetiological agents,clinical features and outcome of septicaemia in infants in Ibadan. WAJM, 22. Bandyopadhyay, S, Bergholte J, Blackwell CD, Friedlander JR, Hennes H. (2002). Risk of serious bacterial infection in children with fever without a source in the post–haemophilus influenzae Era when antibiotics Are reserved for culture-proven bacteremia. Arch Pediatr Adolesc Med, 156:512–517. Bassat, Q., Ogutu, B., Djimde, A., Stricker, K., and Hamed, K. (2015). Tailoring a Pediatric Formulation of Artemether-Lumefantrine for Treatment of Plasmodium falciparum Malaria. Antimicrobial Agents and Chemotherapy, 59(8), 4366-4374. Bauer, A.W., Kirby, W.M.M., Sherris, J.C., Turck, M. (1966): Antibiotic susceptibility testing by a standardized single disk method. Am. J. Clin. Pathol. 45: 493-496. Berger, S. A (1983) Pseudobacteremia due to contaminated alcohol swabs. J Clin Microbiol 18(4): 974-975. Berkley, J, Mwarumba S, Bramham K, Lowe B, Marsh K. (1999). Bacteremia complicating severe malaria in children. Trans. R. Soc. Trop. Med. Hyg. 93:283–286. Berkley, J. A., Bejon, P., Mwangi, T., Gwer, S., Maitland, K., Williams, T. N., and Newton, C. R. (2009). HIV infection, malnutrition, and invasive bacterial infection among children with severe malaria. Clinical infectious diseases, 49(3), 336-343. University of Ghana http://ugspace.ug.edu.gh 49 Berkley, J. A., Maitland, K., Mwangi, I., Ngetsa, C., Mwarumba, S., Lowe, B. S., Newton, C. R. J. C., Marsh, K., Scott, J. A. G., English, M., (2005). Use of clinical syndromes to target antibiotic prescribing in seriously ill children in malaria endemic area. BMJ, 10:1136. Bhandari, A., Dratler, S., Raube, K., and Thulasiraj, R. D. (2008a). Specialty care systems: A pioneering vision for global health. Health Affairs, 27(4), 964-976. Bhandari, P. L., Raghuveer, C. V., Rajeev, A, and Bhandari, P. D. (2008b). Comparative study of peripheral blood smear, Quantitative buffy coat and Modified centrifuged blood smear in malaria diagnosis. Indian J Pathol Microbiol; 51(1): 108-12 Blomberg, B., Manji, K. P., Urassa, W. K., Tamim, B. S., Mwakagile, D. S., Jureen, R., and Maselle, S. Y. (2007). Antimicrobial resistance predicts death in Tanzanian children with bloodstream infections: a prospective cohort study.BMC infectious diseases, 7(1), 1. Bogaert, D., De Groot, R. and Hermans, P. (2004). Streptococcus pneumoniae colonisation: the key to pneumococcal disease. Lancet Infect Dis, 4, 144-154. Boyd, M. F. (Ed.). (1949). Malariology: A comprehensive survey of all aspects of this group of diseases from a global standpoint (Vol. 2). Saunders. Breman, J. G. (2001). The Ears of the Hippopotamus: Manifestations, Determinants, and Estimates of the Malaria Burden. American Journal of Tropical Medicine and Hygiene.;64(Suppl. 12):1–11. University of Ghana http://ugspace.ug.edu.gh 50 Breman, J. G., Alilio, M. S., and Mills, A. (2004). Conquering the intolerable burden of malaria: what’s new, what’s needed: a summary. The American journal of tropical medicine and hygiene, 71(2 suppl), 1-15. Bronzan, R. N., Taylor, T. E., Mwenechanya, J., Tembo, M., Kayira, K., Bwanaisa, L., and Phiri, A. (2007). Bacteremia in Malawian children with severe malaria: prevalence, etiology, HIV coinfection, and outcome. Journal of Infectious Diseases, 195(6), 895-904. Brooks, G. F., Carroll, K. C., Butel, J. S., Morse, S. A. (2007) Medical Microbiology. 24 th Edition. McGraw-Hill Companies, Inc. San Francisco. Pp. 145-150. Brown, H., Rogerson, S., Taylor, T., Tembo, M., Mwenechanya, J., Molyneux, M., & Turner, G. (2001). Blood-brain barrier function in cerebral malaria in Malawian children. The American journal of tropical medicine and hygiene,64(3), 207-213. Bryce, J., Boschi-Pinto, C., Shibuya, K., Black, R. E., & WHO Child Health Epidemiology Reference Group. (2005). WHO estimates of the causes of death in children. The Lancet, 365(9465), 1147-1152. Carpenter, C. C., Pearson, G. W., Mitchell, V. S., & Oaks Jr, S. C. (Eds.). (1991). Malaria: Obstacles and Opportunities. National Academies Press. Carter, R., and Mendis, K. N. (2002). Evolutionary and historical aspects of the burden of malaria. Clinical microbiology reviews, 15(4), 564-594. Caugant, D. A., and Maiden, M. C. (2009). Meningococcal carriage and disease—population biology and evolution. Vaccine, 27, B64-B70. University of Ghana http://ugspace.ug.edu.gh 51 CDC. (2012). CDC Health Information for International Travel 2012. Oxford University Press, New York, USA Cheesbrough, M. (1987). Medical Laboratory Manual for Tropical Countries. Volume 1 (No. Ed. 2). Tropical Health Technology. Chin, W., Contacos, P. G., Collins, W. E., Jeter, M. H., and Alpert, E. (1968). Experimental mosquito-transmission of Plasmodium knowlesi to man and monkey. The American Journal of Tropical Medicine and Hygiene, 17(3), 355-358. Christopher, A., Mshana, S. E., Kidenya, B. R., Hokororo, A., and Morona, D. (2013). Bacteremia and resistant gram-negative pathogens among under-fives in Tanzania. Ital J Pediatr, 39(1), 27. Clinical and Laboratory Standards Institute (CLSI). (2013). M100-S23. Performance standards for antimicrobial susceptibility testing: 23rd informational supplement. CLSI, Wayne, PA. Cook, G.C. and Zumla, A.I. (eds.) Manson's Tropical Diseases (Saunders Elsevier, 2009). Cox-Singh, J., Davis, T. M., Lee, K. S., Shamsul, S. S., Matusop, A., Ratnam, S., ... and Singh, B. (2008). Plasmodium knowlesi malaria in humans is widely distributed and potentially life threatening. Clinical infectious diseases, 46(2), 165-171. Daneshvar, C., Davis, T. M., Cox-Singh, J., Rafa'ee, M. Z., Zakaria, S. K., Divis, P. C., and Singh, B. (2009). Clinical and laboratory features of human Plasmodium knowlesi infection. Clinical infectious diseases, 49(6), 852-860. University of Ghana http://ugspace.ug.edu.gh 52 Del Rio, A., Cervera, C., Moreno, A., Moreillon, P. and Miró, J. M. (2009). Patients at risk of complications of Staphylococcus aureus bloodstream infection. Clin Infect Dis, 48, S246-S253. Dondorp, A, Nosten, F., Stepniewska, K., Day, N., White, N., S. E. A. Q. A. M. A. T. (2005a). Artesunate versus quinine for treatment of severe falciparum malaria: a randomised trial. Lancet; 366:717–25. Dondorp, A. M., Desakorn, V., Pongtavornpinyo, W., Sahassananda, D., Silamut, K., Chotivanich, K., Newton, P. N., Pitisuttithum, P., Smithyman, A. M., White, N. J., Day, N. P. (2005b). Estimation of the total parasite biomass in acute falciparum malaria from plasma PfHRP2. PLoS Med, 2(8), e204. Dondorp, A. M., Fanello, C. I., Hendriksen, I. C., Gomes, E., Seni, A., Chhaganlal, K. D., ... and AQUAMAT group. (2010). Artesunate versus quinine in the treatment of severe falciparum malaria in African children (AQUAMAT): an open-label, randomised trial. The Lancet, 376(9753), 1647-1657. Donkor, E. S., Adegbola, R. A., Wren, B. W. and Antonio, M. (2013). Population biology of Streptococcus pneumoniae in West Africa: multilocus sequence typing of serotypes that exhibit different predisposition to invasive disease and carriage. PloS one, 8, e53925. Dorovini-Zis, K., Schmidt, K., Huynh, H., Fu, W., Whitten, R. O., Milner, D., ... & Taylor, T. E. (2011). The neuropathology of fatal cerebral malaria in malawian children. The American journal of pathology, 178(5), 2146-2158. University of Ghana http://ugspace.ug.edu.gh 53 Eckstein-Ludwig, U., Webb, R. J., Van Goethem, I. D. A., East, J. M., Lee, A. G., Kimura, M., and Krishna, S. (2003). Artemisinins target the SERCA of Plasmodium falciparum. Nature, 424(6951), 957-961. Enwere, G., Biney, E., Cheung, Y., Zaman, S. M., Okoko, B., Oluwalana, C., and Cutts, F. T. (2006). Epidemiologic and clinical characteristics of community-acquired invasive bacterial infections in children aged 2–29 months in The Gambia. The Pediatric infectious disease journal, 25(8), 700-705. Essuman, V. A., Ntim-Amponsah, C. T., Astrup, B. S., Adjei, G. O., Kurtzhals, J. A., Ndanu, T. A., & Goka, B. (2010). Retinopathy in severe malaria in Ghanaian children–overlap between fundus changes in cerebral and non-cerebral malaria. Malar J, 9, 232. Evans, J. A., Adusei, A., Timmann, C., May, J., Mack, D., Agbenyega, T., ... & Frimpong, E. (2004). High mortality of infant bacteraemia clinically indistinguishable from severe malaria. Qjm, 97(9), 591-597. Feasey, N. A., Dougan, G., Kingsley, R. A., Heyderman, R. S., and Gordon, M. A. (2012). Invasive non-typhoidal salmonella disease: an emerging and neglected tropical disease in Africa. The Lancet, 379(9835), 2489-2499. Gallup, J. L., and Sachs, J. D. (2001). The economic burden of malaria. The American journal of tropical medicine and hygiene, 64(1 suppl), 85-96. University of Ghana http://ugspace.ug.edu.gh 54 García, F.,