FUNGAEMIA IN THE NEONATAL UNIT AT CHRIS HANI BARAGWANATH HOSPITAL: RISK FACTORS, AETIOLOGY, SUSCEPTIBILITY TO ANTIFUNGALS AND OUTCOME. FIRDOSE LAMBEY NAKWA STUDENT NO: 9102664T ETHICS CLEARANCE NO: M110848 (PREVIOUSLY M03-10-38) A research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg in partial fulfilment of the requirements for the degree of Master of Medicine in the branch of Paediatrics. Johannesburg, October 2011 ii DECLARATION I, Firdose Lambey Nakwa declare that the research report is my own work. It is being submitted for the degree of Master of Medicine in the branch of Paediatrics in the University of the Witwatersrand, Johannesburg. It has not been submitted before for any degree or examination at this or any other University. _________________________________ ___________day of ______________, 2011. I, certify that the study contained in this thesis has the approval of the Committee for Resarch in Human Subjects of the University of the Witwatersrand, Johannesburg. The ethics clearance number is M110848 (which incorporates the original clearance number M03-10-38) _________________________________ ___________day of ______________, 2011. iii DEDICATION I thank my parents Mohamed Rashid Lambey Nakwa and Zulagha Nakwa, my siblings Mohamed Faried Lambey Nakwa, Zainab Lambey Nakwa, Ismail Lambey Nakwa and Mahmood Lambey Nakwa; and my delightful nephew Rayhaan Lambey Nakwa for their understanding and endless support in my quest to complete this research report. iv PRESENTATIONS Nakwa FL, Velaphi SC, Wadula J, Khoosal MM. Fungal sepsis in a neonatal unit. Conference Proceedings of the Priorities in Perinatal Care; 2006 Mar; Drakensberg, Kwazulu Natal. v ABSTRACT: TITLE: FUNGAEMIA IN THE NEONATAL UNIT AT CHRIS HANI BARAGWANATH HOSPITAL: RISK FACTORS, AETIOLOGY, SUSCEPTIBILITY TO ANTIFUNGALS AND OUTCOME. Aim The aim was to determine the epidemiology of invasive fungal infections at Chris Hani Baragwanath Hospital. The specific objectives were to determine the 1) risk factors, 2) clinical presentation, 3) laboratory abnormalities, 4) organisms and their susceptibilities and 6) outcome in neonates with positive blood or CSF fungal cultures at Chris Hani Baragwanath Hospital. Methods This was a retrospective record review of patients who had positive blood or CSF cultures. Patients were identified by a computerized microbiological surveillance database. The data was collected over a three-year period from January 2002 to December 2004. Patient hospital files were reviewed for clinical signs, full blood count (FBC), C-reactive protein (CRP) and outcomes. Fungal culture results were reviewed for susceptibilities. To identify risk factors a convenient cohort was compared to the patients with fungal sepsis. The data was analysed using a Statistica software package. Results There were 150 patients with fungal sepsis among admissions over this 3 year- period giving an incidence of 1.3 per 100 admissions. Thirty-nine records were not found thus 111 patient records were reviewed. The median birthweight was 1280g vi and the gestational age 30 weeks. The median age of onset was 16 days and 6.3% had early onset fungal sepsis. There were 61 males. Twenty-eight percent of patients were born to HIV positive mothers. Candida parapsilosis was the commonest (56%) organism isolated followed by C. albicans (43%). All the C. albicans isolates and 93% of the C. parapsilosis isolates were susceptible to amphotericin B. Fluconazole susceptibilities were reported as, 96% for C. albicans, and 60% of the C. parapsilosis as being susceptible. Central venous catheters (CVCs) (p=<0.001), the use of TPN (p=<0.001) and third generation cephalosporins were identified as risk factors associated with fungal sepsis. The all-cause mortality and Candida?related mortality were 30% and 23% respectively. The non-survivors had lower platelet counts (p=0.007) than the survivors. Patients with Gram-negative sepsis had lower platelet counts than the fungal group (p=<0.001) on the repeat laboratory parameters. Conclusion The incidence is 1.3 per 100 admissions. Risk factors associated with fungal sepsis are very low birthweight and gestational age, the use of TPN, CVCs and third generation cephalosporins. Candida parapsilosis is the common organism causing fungal sepsis in neonates. Candida albicans was associated with a higher mortality. Thrombocytopenia is not organism specific to fungal sepsis. vii ACKNOWLEDGEMENTS I would like to thank my supervisor Professor SC Velaphi for his tireless assistance and persistence in spurring me on to complete my research report. His guidance and reprisals have shaped me as a researcher and a clinician. Thank you to Dr Kebashni Thandrayen for her statistical support and Ms Zainab Lambey Nakwa for her technical computer support. viii LIST OF TABLES Table 3.1 Characteristics of Infants with Fungal Sepsis 25 Table 3.2 Presenting Signs and Clinical Diagnosis in Patients with Fungal Sepsis 26 Table 3.3 Initial and Repeat Laboratory Findings (FBC and CRP) Among Infants with Fungal Sepsis 27 Table 3.4 Susceptibilities of Fungal Species Isolates to Antifungals 29 Table 3.5 All-cause Mortality of Infants with Fungal Sepsis according to Species 30 Table 3.6 Mortality of Infants with Fungal Sepsis 31 Table 3.7 Comparison of Characteristics Between the Survivors and Non-survivors within the Fungal Sepsis Group for Candida-related Mortality 32 Table 3.8 Characteristics of Patients in the C.albicans group compared to the C.parapsilosis group 34 Table 3.9 Comparison of Initial and Repeat Blood Results Between the C.albicans and the C.parapsilosis group 35 Table 3.10 Comparison of Characteristics of Infants with Fungal to those with Bacterial Sepsis 36 Table 3.11 Comparison of Initial and Repeat Laboratory Parameters Between Infants with Fungal and Bacterial Sepsis 38 Table 3.12 Comparison of the use of Antibiotics, Central lines and TPN Between Infants with Fungal and Bacterial Sepsis 39 ix LIST OF FIGURES Figure 5.1 Fungal species causing sepsis in neonates 28 x LIST OF ABBREVIATIONS ABCD Amphotericin B Colloid Dispersion ABLC Amphotericin B Lipid Complex AIDS Acquired Immunodeficiency Syndrome CRP C-reactive protein CSF Cerebrospinal Fluid CVC Central Venous Catheter ELBW Extremely Low Birthweight EOS Early - onset Sepsis ETT Endotracheal Tube FBC Full Blood Count GIT Gastrointestinal tract GN Gram-negative HCW Health Care Workers HIV Human Immunodeficiency Virus IDSA Infectious Diseases Society of America IgG IL-6 IL-8 Immunoglobulin G Interleukin 6 Interleukin 8 L-AmB Liposomal Amphotercin B LOS Late ? onset Sepsis MDG Millineum Development Goals MIC Minimum Inhibitory Concentration NAC Non-albicans Candida NEC Necrotising Enterocolitis xi NICHD National Institute of Child and Human Development NICU Neonatal Intensive Care Unit R Resistant S Susceptible S-DD Susceptible Dose Dependent TPN Total Parenteral Nutrition USA United States of America UTI Urinary Tract Infection VLBW Very Low Birthweight WHO World Health Organization xii PREFACE Advances in neonatal care have escalated dramatically. With the VLBW infant surviving longer and having prolonged hospital admissions, Candida is emerging as an important pathogen. Candida is associated with a significant morbidity and mortality. This study was undertaken as I had noticed a trend towards an increase in the number of yeasts being isolated in the unit. A study to look at the epidemiology of Candida had not been undertaken in the neonatal unit in the Chris Hani Baragwanath Hospital. Hence, it was imperative to outline the patient characteristics, risk factors, susceptibilities and outcome of the patients with Candida sepsis. The information attained would assist us as clinicians to improve the quality of care and the current management of patients with Candida sepsis xiii TABLE OF CONTENTS DECLARATION II DEDICATION III PRESENTATIONS IV ABSTRACT V ACKNOWLEDGEMENTS VII LIST OF TABLES VIII LIST OF FIGURES IX LIST OF ABBREVIATIONS X PREFACE XII CHAPTER 1 : LITERATURE REVIEW 1 1.1 Background 1 1.2 Candida species causing infections in neonates 2 1.2.1 Candida albicans 5 1.2.2 Candida parapsilosis 5 1.2.3 Candida krusei 7 1.2.4 Candida glabrata 8 1.3 Incidence 8 1.4 Risk Factors 9 1.4.1 Intrinsic Factors 9 1.4.2 Extrinsic Factors 10 1.5 Clinical Presentation and Laboratory Findings 11 xiv 1.6 Treatment 12 1.6.1 Amphotericin B 13 1.6.2 Fluconazole 14 1.6.3 Combination therapy 16 1.6.4 Empiric therapy 16 1.7 Susceptibility of Candida Species to Antifungals 18 1.7.1 Susceptibility to Amphotericin B 18 1.7.2 Susceptibility to Fluconazole 19 1.8 Mortality 20 CHAPTER 2 : OBJECTIVES AND METHODS 21 2.1 Objectives 21 2.2 Methods 21 2.2.1 Study Design 21 2.2.2 Study Population 22 2.2.3 Determining Risk Factors 22 2.2.4 Statistical Analysis 23 CHAPTER 3 : RESULTS 24 3.1 Incidence 24 3.2 Characteristics 24 3.3 Clinical Presentation and Diagnosis 25 3.4 Laboratory Findings 26 3.5 Fungal Isolates 28 3.6 Susceptibilities 29 xv 3.7 Mortality 30 3.8 Comparing Survivors and Non-survivors in the Fungal Group 31 3.9 Comparison Between the C. albicans Group and the C. parapsilosis Group 33 3.10 Risk Factors Associated with the Development of Fungal Sepsis 35 CHAPTER 4 : DISCUSSION 40 CHAPTER 5 : CONCLUSION 47 CHAPTER 6: LIMITATIONS 48 BIBLIOGRAPHY 50 APPENDIX A: PROTOCOL 60 APPENDIX B: DATA SHEET 71 APPENDIX C: ETHICS APPROVAL 74 1 CHAPTER 1 LITERATURE REVIEW 1.1 Background The World Health Organization (WHO) reports that 10 million children under the age of 5 years die each year.1 The mortality rate of children under the age of 5 years in Sub-Saharan Africa is 164 per 1000 and in South Africa is 67 per 1000 live births.2 Neonatal deaths accounts for 30-40% of these under 5 deaths.1 In Sub-Saharan Africa the neonatal mortality rate was reported to be 42-49 per 1000 live births in the year 2000. In the developing world 1.6 million of neonatal deaths are due to neonatal infections.1 One of the Million Development Goals (MDG) is to reduce under-5 mortality rate by 66% by the year 2015.3 Therefore, as part of the process of achieving this goal of reducing neonatal deaths especially those due to neonatal infections; it is important to understand the epidemiology of neonatal infections. This would include identifying the risk factors and causes of these infections. Neonatal infection can be classified as congenital or acquired. In the acquired group it can be sub-divided into early-onset sepsis (EOS) (infections occurring within the first 72 hrs of life) and late-onset sepsis (LOS) (infection occurring after 72 hrs in hospital.4,5 Sepsis is defined as the isolation of an organism from a blood culture taken from a neonate with signs and symptoms of an infection.5,6 The signs of infection are often non-specific during the neonatal period. The diagnosis of sepsis is often made on clinical suspicion. The National Institute of Child and Human Development (NICHD) Neonatal Network and other studies have reported 2 that the most common signs of sepsis are apnoea (55%), gastro-intestinal tract (GIT) symptoms (46%), increased need for oxygen and ventilatory requirements (36%), lethargy and hypotension (25%).5,7-9 The blood culture is the gold standard for the diagnosis of sepsis.4 The blood culture is not as sensitive for fungal elements as it is for bacterial pathogens therefore; fungal elements take longer to culture10. The full blood count (FBC), C-reactive protein (CRP), Interleukin 6 (IL-6), Interleukin 8 (IL-8) and procalcitonin are other useful laboratory tests that are used to assist in diagnosing infection.4,11 Infections due to invasive fungal infections account for 15% of all blood stream infections (BSI) in the neonatal intensive care unit (ICU).12 Candida albicans is the most frequently recovered fungus, but in the last two decades the trend has changed with C. parapsilosis as the more prevalent organism. As invasive candidemia is associated with an attributable mortality of 38% and a crude mortality of 30%-75%; emphasis has been placed on identifying risk factors and empirically treating patients whilst awaiting results.10 In this review and report I am focusing mainly on Candida infections, looking at the pathogens, presenting signs and symptoms, risk factors, treatment and mortality associated with Candida infections. 1.2 Candida species causing infections in neonates Candida is the genus commonly isolated in neonates. Candida organisms are yeasts and exist in a unicellular form. They are ovoid and thin-walled and 3 reproduce by budding.13,14 Candida organisms grow well in blood culture bottles and agar and do not require special fungal media. They form smooth, creamy white glistening colonies and resemble staphylococcal colonies.15 To further identify the organism as a yeast, one can facilitate germ tube formation by placing the organism in serum and within 90 minutes germ tubes are formed. Metabolic tests can be utilized to speciate the organism.15,16 This is based on physiologic parameters rather than morphologic characteristics. They stain as gram positive on haematoxylin and eosin stains. Potassium hydroxide (10%) solution can be used to facilitate hyphae and pseudohyphae formation.15,16,17 Candida species is a normal commensal in humans. It is commonly found on skin, throughout the gastrointestinal tract (GIT), in sputum and the female genital tract.16,17 Candida albicans is found in soil, in animals, in hospital environments, inanimate objects, and food. The non-albicans Candida species live in animal and non-animal environments.18 Candida species isolated from blood or sterile fluid (cerebrospinal fluid, pleural and peritoneal fluid, and urine aspirates) should be considered a pathogen regardless of whether the host is immunocompetent or immunodeficient. Rarely is Candida species a laboratory contaminant. There are more than 150 Candida species but only 15 cause disease in humans.14,19 These include C. albicans, C. parapsilosis, C. glabrata, C. tropicalis, and C. krusei, C. guilliermondii, C. dubliensis, C. rugosa and C. lusitaniae. Candida albicans has been reported in many studies as the commonest Candida species. Candida parapsilosis, C. glabrata, C. tropicalis, and C. krusei are among 4 the common non-albicans Candida (NAC) species. Candida guilliermondii, C. dubliensis, C. rugosa and C. lusitaniae are the less common NAC species.20 Ninety-seven percent (97%) of candidemia is caused by C. albicans, C. parapsilosis, C. tropicalis, C. glabrata, and C. krusei.15,16,18 In neonates the two commonest causes of candidaemia are C. albicans and C. parapsilosis. In the last two decades there has been a decrease in C. albicans and a proportional increase in the non-albicans Candida (NAC) species causing invasive disease.20-22 Kosshoff et al reported a 60% increase in the prevalence of C. parapsilosis over a 5-year period 1991-1995.23 The reasons for the shift in an increase in NAC include the use of prophylactic azoles, the HIV epidemic and the use of fluconazole for oropharyngeal and oesophageal candidiasis.22,24 In Pfaller?s study a geographical as well as an interhospital variation with respect to species isolation and a susceptibility pattern was found.25-27 In the USA and South America, Pfaller reported that C. albicans isolates (53.3%) were more prevalent than NAC (43.8%) differing from Canada where 50% of all the isolates were due to NAC.25 Of the NAC isolates in Canada, C. parapsilosis was the commonest Candida species isolated. There was also a difference in species isolated from different groups of hospitalized patients. In neonates the commonly isolated fungi are C. albicans and C. parapsilosis, whereas patients with haematological malignancies and those with solid tumors are often infected by C. krusei, and C. tropicalis and C. glabrata respectively.14,22,26-28 5 1.2.1 Candida albicans Candida albicans is the most common species (50-70%) isolated in neonates; followed by C. tropicalis, C. parapsilosis and C. glabrata.17,29 It is more virulent than C. parapsilosis. It is a normal commensal of the female genital tract and accounts for most of the early?onset fungaemia in the neonate. It is associated with a higher mortality (44.8%) compared to C. parapsilosis which has been reported to have a mortality rate of 28.5%.30 This is accounted for by its virulence factors. Being a normal commensal of skin it stands to reason that any breach in the skin barrier leads to seeding and the development of invasive disease. It has a cell envelope that has molecular adhesions that bind it to host epithelial surfaces.31 It has a rapid germination upon seeding into the bloodstream, secretes proteinase, form true hyphae, resists phagocytosis and adheres to host surfaces.29 It is an aerobic yeast that can exist anaerobically. It grows into two main forms budding yeast or as continuously extending hyphae with pseudohyphae as an intermediate form. It is naturally diploid and has no known sexual cycle. Candida albicans has been regarded as the dominant Candida species but in the last two decades is being replaced by C. parapsilosis as the common species in the neonatal population. 1.2.2 Candida parapsilosis It is the second most common cause of fungal sepsis in neonates and infants.32,33 The cells are oval, round or cylindrical in shape on Sabouraud?s agar. The colonies are creamy, white, shiny, smooth or wrinkled.34 Candida parapsilosis do 6 not have true hyphae but pseudohyphae and can exist in the yeast form as well. It is a normal commensal of skin and its pathogenicity is limited by an intact skin barrier. It can spread nosocomially by hand carriage of health care workers.35 Its incidence varies with geographical region.27 In the United Kingdom C. parapsilosis accounts for one quarter36 of invasive fungal infections in ill neonates whereas it accounts for one third of bloodstream infections in the United States.37 It has emerged as one of the predominant fungal pathogens in the NICU and is associated with an increased mortality. It can occur without prior colonization; it grows easily in total parenteral nutrition (TPN) and forms biofilms on catheters.34,38,39 Unlike C. albicans, C. parapsilosis organisms are not commonly found in the vaginal tract. The hands of health care workers (HCW) are a major vector coupled with poor hand hygiene and hand washing techniques.27,32,35 The virulence factors include slime production. This causes cell surface hydrophobicity and co- adherence among yeast cells. This leads to aggregation on epithelial cells. They form biofilms on devices and this renders them impermeable to antifungals.27,40 They secrete enzymes e.g. aspartic proteinases that facilitate invasion and colonization of host tissue; phospholipases that disrupts the host membrane, lipases that digest lipids for nutrition and adheres to host cells and tissues.34 7 1.2.3 Candida krusei Candida krusei is one of the NAC that is increasing as a human pathogen. It is isolated from natural habitats and is a transient commensal in man. It has been isolated from mucosal surfaces (oral, GIT, vaginal and anorectal) in healthy individuals. It is commonly isolated from patients who are severely immunocompromised and who have haematological malignancies.41,42 It is one of the three most common yeast combinations isolated with C. albicans in dual Candida infections. The other two being C. glabrata and C. tropicalis. It has the appearance of long grain rice. It grows on Sabouraud?s agar as spreading colonies with a rough, whitish yellow surface. Its cell wall has low reactivity with antisera against other Candida species. It is less virulent than C. albicans and is weakly immunogenic. It is found in two basic forms, yeasts and pseudohyphae and they exist together in growing cultures. C. krusei can grow in vitamin-free media at a temperature of 43-45 degrees Celsius. It is less invasive than C. albicans. Candida krusei do not adhere to and penetrate epithelial cell surfaces as well as C. albicans. It adheres more to inert surfaces than buccal mucosa and therefore has a predilection for implants and catheters. Its biofilm formation is more extensive than any other species because it has increased cell surface hydrophobicity and increased adherence to inert plastic. It does not produce hydrolytic enzymes. Candida krusei is more susceptible to killing by polymorphonuclear cells, macrophages and other immunoeffector cells than C. albicans. 8 1.2.4 Candida glabrata The high prevalence of HIV and AIDS resulting in the long term use of fluconazole has lead to the selection of C. glabrata infection.24,38,40,43,44 Candida glabrata like C. krusei is low in virulence. The lack of hyphae is a factor in this regard. It produces proteinase enzymes and has cell surface hydrophobicity that is similar to C. albicans. It does not possess the B-integrin binding receptor therefore; it is less adherent to epithelial cell surfaces than C. albicans. It does not exhibit phenotypic switching like C. albicans. Host defences do not have to be that stringent in combating C. glabrata because it is less virulent. Risk factors that predispose to C. glabrata infections include prolonged hospitalization and prior antimicrobial use and increased colonization in immunocompromised patients.21 In addition to the other risk factors that predispose to Candida infections, there is an increase in incidence because of prior azole exposure.22,38 It has become an important pathogen because it has decreased susceptibility to antifungal agents.20 It causes infection in any organ and therefore has a diverse clinical picture. C. glabrata candidemia can present with a low grade fever to fulminant septic shock. It has a higher mortality than C. albicans but this is attributed more to host factors than virulence of the organism.44 1.3 Incidence Fungal sepsis is becoming an increasingly important pathogen in the neonatal intensive care units.45 It is the third most common cause of nosocomial bloodstream infection in the United States.14 It affects up to 3000 infants each year in the United States.19,46 9 The frequency has increased by eleven fold in some institutions. In a study by Kossoff et al over a fifteen year period; from 1981 to 1995, the incidence had increased from 2.8% to 28%.23 The NICHD Neonatal Network had also reported that the incidence of all late onset sepsis due to Candida had increased from 9% to 12%.47,48 Chapman has observed the incidence to range from 2.2% to 12.9% in the very low birth weight (VLBW) infant and from 5.5% to 16.5% among the extremely low birth weight infant (ELBW).49,49,50 An overall incidence of 12.3 per 1000 admissions has been reported.7 The overall incidence varies from institution to institution. 1.4 Risk Factors 1.4.1 Intrinsic Factors It has been well documented that very low birth weight infants are at increased risk of developing candidiasis due to the increase in the fungal burden or colonization in this population.4,35,51 The incidence has been reported as 4-15% in the ELBW infant.10,45,51,52 Gestational age has been reported to be inversely associated with Candida sepsis.53 As early as 1986 Bayley et al found that 7.7% of colonized infants developed systemic fungal disease.54 A Taiwanese group documented a 22% fungal colonization rate within two weeks after birth and reported that one fifth of colonized infants will develop invasive disease. This is another risk factor for fungal sepsis.7,55 Host factors such as immaturity of the immune system and the 10 infant?s fragile skin and mucous membranes play a role in the development of fungal sepsis.35 The mucous membrane defenses are decreased in the VLBW infant. The breakdown of skin through the insertion of the central venous catheter per se is an independent risk factor.14 Firstly there is a breach of the skin. Secondly, the organism forms a biofilm which surrounds the catheter. The Candida organism harbors itself within the biofilm thus protecting itself from the antifungals. The neonate has an altered innate or adaptive immunity.14 Serum complement levels, fibronectin, and defenses and cytokine production are decreased in all newborns; in premature babies to a greater degree than the term infants.18 There is a limited production of antibodies in response to the invading pathogen. Cellular defenses such as chemotaxis, phagocytosis and microbiological killing are relatively impaired.11 The use of intralipid as a feeding modality further decreases the chemotactic ability of the neonatal leukocytes. This renders the premature infant more susceptible to infections and leads to, dissemination and deep-seated tissue infections.7 Maternal IgG transfer usually occurs in the third trimester and confers some immunity on the neonate.11,56 Thus, in a premature baby there is a limitation on passively acquired immunity.37 1.4.2 Extrinsic Factors The other risk factors include the use of H2 blockers, indomethacin, steroids and antibiotics especially carbepenems and third generation cephalosporins.4,57,58 The introduction of foreign bodies like the use of instrumentation such as endotracheal 11 intubation (ETT), central venous catheters (CVCs), umbilical venous or arterial catheters and nasogastric tubes in sick neonates are other risk factors.17,35,52,53,59 The length of stay in the neonatal unit exposes these infants to nosocomial bacterial infections therefore increasing the need for venous access to administer antibiotics especially third generation cephalosporins.31,60 Feeding intolerance and the use of intralipid solutions make these infants more susceptible to candidemia.40,61 Total parenteral nutrition (TPN) mainly intralipid acts a good medium for fungal elements to grow. Abdominal surgery has also been identified as a risk factor.32,51,52 1.5 Clinical Presentation and Laboratory Findings Patients with invasive Candida infections can present with non-specific signs of infection. The clinical manifestations range from temperature and glucose instability, respiratory distress, increase in oxygen requirements, features of necrotizing enterocolitis (NEC), feeding difficulties, lethargy, abdominal distension and vomiting.17 This makes it difficult to identify whether the offending pathogen is a fungal organism or a bacterial organism. This often results in delays in starting the appropriate treatment.57 Among the laboratory parameters a high C-reactive protein and thrombocytopenia62 are features that have been identified to be associated with Candida infections.17,57 Positive blood cultures and cultures from other sterile sites are a gold standard to diagnose invasive Candida infections.17. However, blood culture has a low yield; between 50-80% for a positive fungal result.14,19,57 The 12 yield on the blood culture also takes a long time, as the organism is slow growing and may take a while before it becomes positive.37 At least 30 % of neonates are diagnosed with invasive candidiasis at post mortem. Newer methods to diagnose fungal infections have been formulated and are currently being tested. This includes B-D glucan test, breath test, and growth on a special broth. Because of the difficulties in making a diagnosis of fungal sepsis, Benjamin et al has developed a scoring system looking at the gestational age and other risk factors. The clinical predictive score is based on the probability of developing fungal infection when a patient has these risk factors.10 A score of 0, 1 or 2 is assigned based on risk factors such as decreased gestational age, thrombocytopenia and the use of third generation cephalosporins. If the neonates were thrombocytopenic a score of 2 was assigned (otherwise 0), if they received a 3rd generation cephalosporins or a carbepenem a score of 1 was assigned (otherwise 0), if the gestational age was 25-27 weeks a score of 1 was assigned (otherwise 0), if the gestational age was <25 weeks a score of 2 was assigned. If the combined score was 2 or more it was regarded as a positive candidemia score. A score >2 has a sensitivity of 85% and a specificity of 47%. Once a patient meets these criteria antifungal treatment would be started empirically. 1.6 Treatment The treatment modalities are limited but newer drugs are being formulated and there are some trials that are being conducted. The treatment is limited and is fraught with its own complications depending on the organism?s sensitivity 13 patterns, the patient?s medical condition and the epidemiology of the strains in the NICU. Polyenes (Amphotericin B), azoles (fluconazole, voriconazole, itraconazole, miconazole) and fluorinated pyrimidines (flucytosine), echinocandins (caspofungin, micafungin) are among the armamentarium used to treat candidemia.17,50,63 The antifungal agents most commonly used are amphotericin B and fluconazole, therefore the next two sections will review these two drugs. 1.6.1 Amphotericin B The gold standard of therapy for candidiasis is amphotericin B.17 It was first formulated in the 1950s. Amphotericin B is fungicidal. It binds to the ergosterol component of the fungal cell wall; this then leads to membrane leakage and eventually cell death.50,64 Its use is limited by its safety profile. It causes nephrotoxicity and hepatotoxicity. It is often used empirically for suspected systemic candidiasis.63 It is sometimes used in combination with fluconazole for treating refractory candidiasis.65 Some studies show antagonism between the two drugs but clinically these drugs are synergistic with a positive reponse.63,66,67 It is also used in combination with flucytosine to treat meningitis due to candidiasis.65 Amphotericin B comes in different forms namely Amphotericin B lipid complex (ABLC), Amphotericin B colloid dispension (ABCD), liposomal Amphotericin B (L- AmB). These lipid formulations have similar efficacy and a better side-effect profile than the standard amphotericin B.30,68-70 The other advantage is that higher doses 14 can be used.71,72 Their use has been advocated in renal failure and when there is failure of standard amphotericin B treatment.30,50,73,74 Lipid formulations are better if given as first line antifungal treatment and at high doses.17 Study by Linder has demonstrated a shorter duration of antifungal therapy with lipid formulations and that there is a better eradication of the organism.60 Liposomal AmB has the greatest renal protection of the three lipid formulations.63,70 Despite its high side- effect profile, amphotericin B is the gold standard of treatment for invasive candidiasis.63,75 1.6.2 Fluconazole Fluconazole is an azole that was developed in the 1970s. It acts by inhibiting CYP- dependent 14-?-demethylase.50 This inhibition interrupts the conversion of lanosterol to ergosterol. Ergosterol is a major component of the fungal cell wall.64 This then increases membrane permeability and leads to cell leakage and eventually cell death. It also leads to cell growth inhibition, morphologic changes, cessation of sterol synthesis, and reduction in adhesion to epithelial cells. Fluconazole is fungistatic and not fungicidal. It has good penetration into cerebrospinal fluid (CSF), liver, kidney and spleen. It is excreted unchanged in the urine therefore it is a good agent for treating UTI (urinary tract infections). Fluconazole is as effective as amphotericin B and it has less toxicity therefore is a safer alternative to amphotericin B in treating candidiasis.76 Driessen et al has reported that fluconazole has the same efficacy in clearing the infection as amphotericin B.77 Some centers use fluconazole as a first line agent as empiric therapy because of its low side effect profile and excellent oral bioavailability.78 15 Most Candida species are susceptible to fluconazole, but C. glabrata is dose- dependent and C. krusei is resistant.78-80 However, resistance to fluconazole has been reported. This is due to the HIV epidemic which has resulted in the widespread use of fluconazole to prevent and treat candidiasis. This has lead to more non-albicans Candida (NAC) species such as C. glabrata and C. guilliermondii becoming resistant to fluconazole.21,40,43 Fluconazole has been shown to reduce the incidence of colonization and invasive candidemia when used as a prophylactic agent.81-84 There has been a concern in the emergence of Candida strains (C. krusei and C. glabrata) that exhibit resistance to fluconazole.85 This has been thought to be due to the use of fluconazole for prophylaxis in neonates at risk of invasive fungal sepsis. Many studies have documented no resistance to fluconazole in patients that have received prophylaxis.81-83 Kauffmann et al has speculated that with the lower dose and less frequent dosing schedule resistance may be prevented.83 However, resistance can develop over time in a population that has received fluconazole.60,86 A recent Cochrane meta-analysis has shown a decrease in fungal infections with fluconazole prophylaxis in VLBW infants.87,88 Recommendations are that high risk populations should receive prophylaxis. The low dose and shorter dosing schedule should preclude from the development of resistant strains. Caution should still be exercised as the threat of resistance is ever looming. 16 Antifungal prophylaxis decreases the fungal burden and thereby decreases the incidence of colonization and invasive candidiasis.89 Manzoni and others had shown that this strategy is effective in decreasing invasive Candida infections in the neonatal population.84 However, the development of NAC strains and resistant Candida strains is still an ever looming problem.21 1.6.3 Combination therapy Amphotericin B and flucytosine has been the first combination used for meningitis, endocarditis, and peritonitis caused by Candida.90 The use of fluconazole and amphotericin B has shown antagonism in vitro but clinically these drugs are synergistic with a positive response.63,65 A study by Rex et al has shown favourable outcomes.91 There was a success rate of 68% vs 56% when amphotericin B and fluconazole were used in combination than when fluconazole used alone. The mortality was the same in both groups. Thus, amphotericin B and fluconazole are used in combination to treat refractory candidiasis. There is very limited data on the use of echinocandins in combination with amphotericin B.65 It should have an additive effect and synergistic effect as caspofungin inhibits the cell wall synthesis and enhances the access of amphotericin B to its target which is the cell membrane.92 1.6.4 Empiric therapy Fungal sepsis is a leading cause of mortality and morbidity but this is organism specific. Empiric therapy has been shown to decrease the mortality and morbidity 17 in at risk patients. Most of the studies have been in adult cancer patients with a febrile illness and neutropenia.93 The choice of antifungals range from amphotericin B to voriconazole. Amphotericin B is the most efficacious empiric antifungal used but it is quite toxic.94 Comparative studies have shown that the efficacy decreases with the other classes of antifungals but the side effect profile improves. Kaufman and Fairchild have identified high-risk ELBW infants with risk factors such as CVCs, ETT in-situ, platelet count <100 000/mm3, broad spectrum antibiotics, exposure to carbepenems and cephalosporins and a gestational age less than 28 weeks as a subset of patients that would benefit from empiric therapy.7,38,95,96 A Candida clinical predictive model has been devised to assist in commencing empiric therapy as fungal elements take a long time to grow.10 The duration of empiric treatment would be for as long as the blood culture is formally interpreted as negative i.e. if a blood culture is regarded as negative after seven days then empiric therapy would be continued for seven days.7 In some institutions, if a patient is not responding to antibacterial treatment, and has one or more of the risk factors and/or features suggestive of a fungal infection; antifungal therapy would be commenced empirically. The choice of drug would depend on the epidemiology of Candida isolated in the NICU, the patient population and the medical condition of the patient10 and whether the patient received fluconazole prophylaxis or not. If fluconazole prophylaxis is practiced in the unit then an antifungal other than fluconazole should be commenced as empiric treatment. Cross resistance as well as co-resistance to azoles are confounding variables. Non-albicans Candida isolates are also emerging as important pathogens21 thus, the choice an antifungal as empiric therapy may be difficult. 18 1.7 Susceptibility of Candida Species to Antifungals 1.7.1 Susceptibility to Amphotericin B Susceptibilities are carried out by detecting the minimum inhibitory concentrations (MIC) of the particular organism to an antifungal.97 The MIC is defined as the lowest concentration of amphotericin B and fluconazole to reduce the turbidity of cells to >95% and 50%. For fluconazole if the MIC < 8ug/l the organism is classified as susceptible(S); and if the MICs are > 64ug/l the organism is classified as resistant (R). If the organism has an MIC between 16 and 32 ug/l it is classified as susceptible dose-dependent (S-DD), which means that at a higher dose of the antifungal the organism would be rendered dead. In the case of amphotericin B an MIC of <1ug/l is susceptible (S) and resistant as an MIC >2 ug/l.96,97,98 Ninety-five percent (95%) of C. albicans isolates are susceptible to amphotericin B, whilst 97-98% of C. parapsilosis isolates are susceptible.95,99 Clerihew has reported that all C. parapsilosis isolates in her study are sensitive to amphotericin B.36 Candida glabrata and C. krusei isolates are relatively resistant to the amphotericin B.25,95 These two organisms have higher MICs for amphotericin B67 and thus, the IDSA guidelines recommends a higher dosage of amphotericin B when treating them.63,66 Yang and Shoa in Taiwan and Almirante in Spain have reported similar findings.79,100,101 It has also been reported that there is an emerging resistance of C. tropicalis and C. guillermondii to amphotericin B.38,99,102, whilst C. lusitaniae isolates are resistant to amphotericin B.78 19 1.7.2 Susceptibility to Fluconazole Most C. albicans isolates are sensitive to fluconazole103,104 whereas the NAC tend to have higher resistance rates to fluconazole.25,43,85,102,105 In China Ying Liang Yang et al found that all C. parapsilosis isolates were sensitive to fluconazole and the C. krusei isolates having had the highest resistance rate to fluconazole.79,106 Nine percent (9%) of C. glabrata and 100% of C. krusei isolates have been reported to be resistant to fluconazole.25,31,104 Newer triazoles have been developed but the C. glabrata isolates had shown cross resistance to the triazoles hence, amphotericin B was used as therapy for this isolate. With the increased use of amphotericin B the resistance rates of C. glabrata isolates were reported as high as 20-36% in North America. Co-resistance to amphotericin B and fluconazole by this isolate may be an impending problem.79 Thus, it is wise to monitor surveillance patterns so that emerging co-resistance can be identified and an appropriate antifungal is chosen as empiric therapy. Azole prophylaxis has been implicated as causing a shift towards an increase in C. krusei and C. glabrata infections.21,100,103,107 These strains are known to be inherently resistant to fluconazole. Whilst the overall incidence of candidemia is decreasing as azole prophylaxis decreases the rate of C. albicans infections there is an increase in infections with NAC species.40 There should be an awareness when using fluconazole as empiric therapy in centers where fluconazole is used as a prophylactic agent.104 20 Sarvikivi has shown the emergence of a resistant C. parapsilosis strain to fluconazole.86 This had occurred in patients in Finland that received fluconazole prophylaxis at the time that the blood culture was taken. This is in contrast to other studies that have proven that C. parapsilosis remains sensitive to fluconazole despite the patients having received prophylaxis. Therefore one needs to bear in mind that resistance can develop over time. 1.8 Mortality The mortality rate is 20 ? 40% and it depends on the Candida species.8,9,14,47,51,57,63 The highest mortality is reported as being due to C. albicans among those with fungal sepsis.30,47,75 This was noted by Faix in a single centre study with C. albicans accounting for 24% of the deaths and while there were no deaths in those with C. parapsilosis.108 The NICHD Neonatal Network had also reported a significantly higher mortality (44%) due to C. albicans compared to 16% of those with C. parapsilosis.47 This is also attributed to the timing of the infection. Candida albicans is vertically transmitted causing sepsis early in the more compromised host and C. parapsilosis causing infection in the older more immune competent host.7-9,109 The crude mortality of C. glabrata and C. krusei is reported as 50% and 100% respectively110. In a study by Pappas and Rex the Candida- specific mortality in patients <13years for C. glabrata was 13% for Candida- specific mortality and 0% for C. krusei. This study also found that the survival rates were not any worse for C. glabrata isolates as compared to the other Candida species. If the patients with C. glabrata isolates were not treated it was associated with a lower survival rate compared to those who received treatment.111 21 CHAPTER 2 2. OBJECTIVES AND METHODS 2.1 Objectives The objective for this study was to determine the epidemiology of invasive culture proven fungal infections at Chris Hani Baragwanath Hospital. The specific objectives were to determine 1. the risk factors associated with fungal sepsis 2. clinical presentation of neonates with fungal sepsis 3. laboratory abnormalities in infants with fungal sepsis 4. the Candida species causing infections in neonates, 5. Susceptibilities of Candida species to commonly used antifungal agents 6. the mortality rates among neonates infected with Candida 2.2 Methods 2.2.1 Study Design This was a retrospective record review. Patients were identified by a computerized microbiological surveillance database. Once identified the patient records were retrieved and analyzed. The data was collected over a three year period from January 2002 to December 2004. To identify the risk factors a control group was a convenient sample which was from records of patients who had culture proven Gram negative (GN) sepsis collected over the same time period. 22 2.2.2 Study Population Patients who had positive blood and /or CSF cultures were identified through looking at the laboratory database. Patient?s hospital records were retrieved for data collection. The data was collected with the aid of a data collection sheet (Appendix A). This sheet included section 1 which concentrated on both infant and maternal demographic data; section 2 dealt with the infective demographics and section 3 was dedicated to the infants? presentation at the time of sepsis. This included details such as the laboratory parameters full blood count (FBC), C- reactive protein (CRP) and blood culture, and cerebrospinal fluid (CSF) results both at the time of presentation and repeat results. Repeat results were the very next laboratory parameters (FBC, CRP blood and CSF culture) sampled from the same patient around the time of infection. It also included details of antibiotic usage, the use of central lines, and total parenteral nutrition. The details of the organism and its susceptibilities were also collected. Data on the outcome that is whether the patient died secondary to fungal sepsis was collected. Death due to fungal sepsis was defined as a patient having demised within two weeks of the positive fungal culture. Antifungals that were used in the unit during this period were amphotericin B and fluconazole. Ethics approval was obtained from the University of the Witwatersrand Committee for Research of Human Subjects. 2.2.3 Determining Risk Factors To determine the risk factors for the development of fungal sepsis, data from infants who were infected with Gram-negative organisms were used for 23 comparison. The Gram-negative cohort was a convenient sample as data had already been collected for another study. 2.2.4 Statistical Analysis The information was entered onto a computer database and analyzed using the Statistica statistical software package. Mean and standard deviations were used to describe data with normal distribution and medians and 25th and 75th percentiles were used to describe data without normal distribution. Comparison between categorical variables was performed using the Chi-squared or Fisher exact test. Comparison between continuous variables was performed using a Students? t-test when there was a normal distribution otherwise a Mann-Whitney U test was used. A significant difference was taken as a p-value of less than 0.05. 24 CHAPTER 3 3. RESULTS 3.1 Incidence One hundred and fifty (150) patients were identified to be infected by Candida from January 2002 to December 2004. There were 11589 admissions to the neonatal unit over this three year period. This gave an incidence of 1.3 cases of Candida per 100 admissions. Patient hospital records of 39 patients were not found leaving 111 infants for analysis. 3.2 Characteristics Characteristics of the 111 infants that had their hospital records reviewed are shown in Table 3.1. Ninety percent of the patients reviewed were low birthweight (LBW) and premature, with a median birthweight of 1280g and a gestational age of 30 weeks. Seventy-three percent (81/111) of the patients were VLBW (weighing<1500g). Fifty-nine percent (59%) of the patients were born vaginally and in 93 patients (84%) the maternal HIV status was known. Of those with known HIV status 28% were born to mothers who were HIV positive. The median age of onset of patients who had positive Candida cultures was 16 days, and 6.3% of patients were less than 3 days old. 25 Table 3.1 Characteristics of Infants with Fungal Sepsis Birth Weight (g)* 1280 (660-3300) Gestational Age (weeks)* 30 (23-40) Mode of delivery Vaginal 66 (59%) Abdominal 35 (32%) Maternal HIV Status Positive 31 (28%) Negative 62 (56%) Unknown 18 (16%) Apgar 1 minute 7 (1-10) 5 minutes 8 (4-10) Gender (n=111) Female 49 (44%) Male 61 (55%) Not specified 1 (1%) Age at onset (days) 16 (1-86) * Median (Ranges) 3.3 Clinical Presentation and Diagnosis The common presenting signs were related to the respiratory system. The three most common presenting symptoms were an increase in the ventilatory requirements, temperature instability and abdominal distension (Table 3.2). Most of the patients were diagnosed as having candidemia as there was no specific clinical diagnosis given. These patients had clinical signs and a positive Candida culture. The common clinical diagnosis was NEC (35%). This is a primary diagnosis. The patients presented with NEC and subsequently developed fungal sepsis. 26 Table 3.2 Presenting Signs and Clinical Diagnosis in Patients with Fungal Sepsis n (%) Signs General (n=49) Temperature instability 42 (22) Lethargy 5 (2.6) Gastrointestinal tract (GIT) (n=42) Abdominal distension 22 (11) Nasogastric aspirates 14 (7) Vomiting 6 (3) Respiratory (n=84) Apnoeas 21 (11) Respiratory distress and/or desaturating 63 (33) Metabolic (n=18) Hyperglycaemia 13 (6.6) Hypoglycaemia 5 (2.6) Central nervous system (CNS) (n=2) Seizures 2 (1) Diagnosis (n=111) Candidemia 54 (49) Necrotizing Enterocolitis 39 (35) Nosocomial Pneumonia 14 (13) Meningitis 7 (6) 3.4 Laboratory Findings Most patients (98%) had a full blood count (FBC) and CRP done as part of the work-up at the time of sepsis. The majority (82%) of patients had a normal white cell count (WCC) with only 7(6%) having leucopenia (WCC <5 x 109/l) and 13 (12%) having a leucocytosis (WCC >25x109/l) (Table 3.3). Fifty-six percent of patients had a thrombocytopenia (platelet count <150x109/l). Seventy-three percent of patients had an abnormal CRP with 66% having a CRP of > 20mg/l. 27 Table 3.3 Initial and Repeat Laboratory Findings (FBC and CRP) Among Infants with Fungal Sepsis Initial Repeat n=109 n = 99 White cell count* 13.3 (1.5-57.1) 12.8(2.6-43.5) Leucopenia (< 5 x 109/l) 7 (6%) 10 (10%) Normal ( 5-25 x 109/l) 89 (82%) 76 (77%) Leucocytosis (>25 x 109/l) 13 (12%) 13 (13%) n=109 n = 99 Platelet count* 123 (5-755) 92(4-721) Thrombocytopenia (<150x 109/l) 61 (56%) 62 (63%) Normal count (150 ? 450 x 109/l) 41 (38%) 31 (31%) Thrombocytosis (>450 x 109/l) 7 (6%) 6 (6%) n=107 n = 88 C-reactive protein* 45 (1-542) 30.5(1-209) Normal <10 mg/l 30 (27%) 18 (20.5%) Borderline (10-20 mg/l) 8 (7%) 18 (20.5%) Increased >20 mg/l 72 (66%) 52 (59%) * Median (Ranges) Ninety-nine (89%) of the 111 patients had a repeat FBC and 88 (79%) patients had a repeat CRP. Repeat FBC and CRP were requested at the discretion of the clinicians and are usually done when there is an abnormal count or if the patient has not improved clinically. The median number of days for both the FBC and CRP to be repeated was 3 days. Comparing the first FBC and the repeat FBC the number of patients with an abnormal WCC increased from 18% to 23% and the number of those with thrombocytopenia increased from 56% to 63% with the median platelet count decreasing from 123x109/l to 92x109/l. The number of 28 patients with a high CRP increased from 73% to 79%. Overall there was a trend towards a decrease in platelet count and WCC, and an increase in the CRP. 3.5 Fungal Isolates The common isolates identified among neonates with fungal sepsis were Candida parapsilosis (63) and Candida albicans (43) (Figure 5.1). Only 6 were caused by other species, 4 by C. glabrata and 2 by C. krusei. One patient had isolated two fungal species during the same episode of fungal sepsis that is a C. albicans and C. glabrata. The organisms were not cultured from the same blood specimen but cultured within 10 days of each other. The patient was treated with a combination of amphotericin B and fluconazole and survived. The combination therapy was used as C. albicans was repeatedly isolated from 3 cultures done at different times. Figure 5.1 Fungal species causing sepsis in neonates 56% 38% 4% 2% C. parapsilosis C. albicans C. glabrata C. krusei 29 3.6 Susceptibilities Among the commonly isolated species susceptibilities were available for 25 of the 43 (58%). C. albicans isolates for both amphotericin and fluconazole and 41 of the 63 (63%) C. parapsilosis isolates were available for amphotericin B and 40 of the 63 (63%) for fluconazole (Table 3.4). One of the C. parapsilosis isolates did not have susceptibilities documented for fluconazole. Among the C. albicans species with susceptibility results available 100% were sensitive to amphotericin B and 96% sensitive to fluconazole. Whereas for C. parapsilosis the susceptibility patterns among those with available susceptibility results were 93% for amphotericin B and 58% for fluconazole. Table 3.4 Susceptibilities of Fungal Species Isolates to Antifungals Organism (n) Amphotericin B Fluconazole C. albicans (25) Susceptible 25(100%) 24 (96%) Intermediate 0 (0%) 0 (0%) Resistant 0 (0%) 1 (4%) C. parapsilosis (41) Susceptible 38(93%) 23 (58%) Intermediate 2 (5%) 9 (23%) Resistant 1 (2%) 8 (20%) C. glabrata (3) Susceptible 1 (33%) 3 (100%) Intermediate 2 (67%) 0 (0%) Resistant 0 (0%) 0 (0%) 30 3.7 Mortality Among the 111 patients that were reviewed 27 died before hospital discharge giving an all-cause mortality rate of 24% for patients diagnosed with Candida (Table 3.5). Twenty-one patients died within 14 days of being diagnosed with Candida; giving a case fatality rate of 19% (Table 3.6). Case fatality rates for C. albicans and C. parapsilosis were high at 23% and 16% respectively. Only 1 patient among the 6 patients who were infected by C. glabrata and C. krusei demised. Table 3.5 All ?Cause Mortality of Infants with Fungal Sepsis according to Species Organism Number All cause mortality All-cause mortality C. albicans 43 13 30% C. parapsilosis 63 13 20% C. glabrata 4 0 0% C. krusei 2 1 50% Total 112* 27 24% * one patient had cultured 2 organisms 31 Table 3.6 Mortality of Infants with Fungal Sepsis Organism Number Candida - related mortality Case fatality rates C. albicans 43 10 23% C. parapsilosis 63 10 16% C. glabrata 4 0 0% C. krusei 2 1 50% Total 112* 21 19% * one patient had cultured 2 organisms 3.8 Comparing Survivors and Non-survivors in the Fungal Group Table 3.7 is a comparison between the characteristics of the survivors and non- survivors in the fungal group for the deaths related to Candida sepsis. Comparing the survivors and non-survivors, the only difference noted was that the non- survivors were older (18 days vs 15 days, p=0.04) and had lower platelet counts (39x109/l vs 151x109/l, p=0.007) than the survivors at the time of onset of sepsis. There were no significant differences in the birthweight, gestational age, gender and mode of delivery and maternal HIV status identified between the two groups. 32 Table 3.7 Comparison of the Characteristics Between the Survivors and Non- Survivors within the Fungal Sepsis Group for Candida-related mortality Survivors Non-survivors p-value n=81 n=21 Birth Weight (g)* 1300 (660- 3300) 1145 (820-3200) 0.7 Gestational age (weeks) * 30 (23 ? 39) 29 (26- 40) 0.6 Booked 0.104 Yes 69 14 No 9 5 Unknown 3 2 Gender 0.245 Female 38 7 Male 42 14 Ambiguous 1 Mode of Delivery 0.492 Vaginal 49 11 Abdominal 27 5 Maternal HIV status 0.87 Positive 25 6 Negative 47 7 Unknown 9 1 Age at onset (days)* 15 (1 ? 60) 18 (2 -86) 0.04 Initial n=80 n=21 WCC X 109/l* 12.45 (1.50-57.1) 13.5 (3.5-37.8) p=0.990 Platelet count X109/l* 151.50 (5-755) 39 (5 ? 422) p=0.007 CRP mg/l* 37.2 (1- 542) 49 (2.2 -130 p=0.821 Repeat n=73 n=18 WCC X 109/l* 11.7 (2.69-37.4) 13.3 (4.4- 36.7) p=0.423 Platelet count X109/l* 132.0 (9.0-721) 28 (4.0-331) p=0.002 n=14 n=66 CRP mg/l* 23.3 (1.00-209) 79.4 (4.0 ? 200.2) p=0.013 *Median (Ranges) 33 3.9 Comparison Between the C. albicans Group and the C. parapsilosis Group In comparing the infants who were infected by C. albicans to those with C. parapsilosis, the only difference was that those that were infected with C. parapsilosis were smaller (p=0.022) and had a lower gestational age (p=0.002) than those infected with C. albicans (Table 3.8). 34 Table 3.8 Characteristics of Patients in the C. albicans group compared to the C. parapsilosis group. Characteristic C. albicans C. parapsilosis p-value n= 43 n= 63 Birthweight (g)* 1594.65 (850-3200) 1335.80 (660-3300) 0.022 Gestational age (weeks)* 32.20 (26.0-40.0) 30.04 (23.0-39.0) 0.002 Booking status 0.848 Yes 35 50 No 7 9 Unknown 1 4 HIV status 0.119 Positive 15 16 Negative 18 39 Unknown 10 7 Mode of delivery 0.136 Vaginal 29 23 Abdominal 10 34 Gender 0.272 Male 21 37 Female 22 25 Ambiguous 1 Apgar 1 minute* 6 (1-10) 6 (1-9) 0.962 5 minute * 8 (4-10) 7 (4-10) 0.443 Age at onset (days)* 20 (2-86) 18 (1-66) 0.61 Death 0.316 Yes 10 10 No 29 47 *Median (Ranges) 35 Table 3.9 Comparison of Initial and Repeat Blood Results Between the C. albicans and the C. parapsilosis Group C. albicans C. parapsilosis p-value Initial n=41 n=63 WCC x109/l* 13.0 (1.50-30.9) 14.8 (3.5-37.8) 0.61 n=41 n=63 Platelet count x109/l* 136.8 (5.0-422.0) 179.7 (13.0-755.0) 0.162 n=43 n=62 CRP mg/l 63.7 (2.0-191.0) 53.5 (1.0-542.0) 0.463 Repeat WCC x109/l* n=38 n=56 15.8 (8.3-22.1) 13.2 (8.2-15.6) 0.143 n=38 n=56 Platelet count x109/l* 147.9 (27.0-210.0) 144.2 (32.0-232.0) 0.912 n=29 n=54 CRP mg/l 62.2 (18.5-101.2) 40.9 (11.0-49.0) 0.031 *Median (Ranges) There was no organism specific difference between the initial and repeat FBC with respect to the white cell count and platelets. However, there was a significant difference in the repeat CRP between the two groups with the C. albicans group having a much higher CRP count (p=0.031) (Table 3.9). 3.10 Risk Factors Associated with the Development of Fungal Sepsis To determine the risk factors associated with development of fungal sepsis; a comparison was made between the study group and a convenient sample of infants infected with Gram-negative (GN) organisms. There was a significant 36 difference in birthweight, gestational age and day of life of onset of sepsis between the two groups. The fungal sepsis group weighed less (p=0.002), and had a lower gestational age (p=0.03) but a later day of onset of sepsis (p=<0.01) and had lower Apgar scores at 1 min (p=0.044) and 5 min (p=0.048) when compared to the GN group (Table 3.10). Table 3.10 Comparison of Characteristics of Infants with Fungal Sepsis to those with Bacterial Sepsis. Characteristic Fungal Bacterial p-value n=111 n=112 Birthweight * 1280 9660-330) 1480 (780-3980) 0.002 Gestational Age* 30 (23 - 40) 32(25-42) 0.003 Day of life onset* 16 (1-86) 8 (1-90) <0.001 Apgar 1 minute* 7 (1-10) 8 (1-10) 0.044 5 minutes* 8 (4-10) 9 (3-10) 0.048 Maternal HIV status n = 93 n=104 0.156 Positive 31(33%) 46 (44%) Negative 62 (67%) 58 (56%) Gender 0.303 Female 49 44 Male 61 75 Mode of delivery 0.677 Vaginal 66 82 Abdominal 35 37 Deaths related to sepsis 21 (19%) 28 (23%) 0.670 *Median (Ranges) In comparing the laboratory results the WCC was significantly lower (p<0.001) in the bacterial group in both the initial and repeat blood results. The platelet count and CRP were similar between the two groups at the time of the sepsis work-up, however, on the repeat FBC; the platelet count was significantly lower 37 (p=0.013) and the CRP (p=0.013) was significantly higher in the bacterial group. There was no statistically significant difference in mortality rate due to sepsis between the two groups (Table 3.11). 38 Table 3.11 Comparison of Initial and Repeat Laboratory Parameters Between Infants with Fungal and Bacterial Sepsis Fungal Bacterial p- value Initial blood results n = 109 n = 120 White cell count* 13.3 (1.5 -57.1) 8.8( 0.8-33.3) <0.001 Leucopaenia (<5x109/l) 7 (6%) 32 (27%) Normal WCC (5?25x109/l ) 89 (82%) 85 (71%) Leucocytosis (>25x109/l)) 13 (12%) 4 (3%) n = 109 n = 120 Platelet count* 122 (5-755) 97 ( 5-510) 0.181 Thrombocytopaenia (<150x109/l) 61 (56%) 78 (65%) Normal (150?450x109/l) 41 (38%) 41 (34%) Thrombocytosis (>450x109/l) 7 (6%) 1 (1%) n = 107 n = 100 C-reactive protein* 43(1-542) 33.5 (1-314) 0.641 Normal (<10 mg/l) 30 (27%) 27 (27%) Borderline (10-20 mg/l) 8 (7%) 14 (14%) Increased (>20 mg/l) 72 (66%) 59 (59%) Repeat blood results n = 99 n = 73 White cell count* 12.8(2.6-43.5) 8.6 (1.19-31.5) <0.001 Leucopaenia (<5 x 109/l) 10 (10%) 17 (23%) Normal WCC (5 ? 25 x 109/l ) 76 (77%) 49 (67%) Leucocytosis (>25 x 109/l)) 13 (13%) 7 (10%) n = 99 n = 72 Platelet count* 92 ( 4-721) 45 (7 -393) 0.013 Thrombocytopaenia (<150x109/l) 62 (63%) 58 (81%) Normal (150? 450x109/l) 31 (31%) 14 (19%) Thrombocytosis (>450x109/l) 6 (6%) 0 (0%) n = 88 n = 40 C-reactive protein* 30.5(1-209) 72 ( 1-296) 0.013 Normal (<10 mg/l) 18 (20.5%) 5 (12.5%) Borderline (10-20 mg/l) 18 (20.5%) 5 (12.5%) Increased (>20 mg/l) 52 (59%) 30 (75%) *Median (Ranges) 39 There were more patients on antibiotics other than Penicillin G and Gentamicin (p=<0.001), who had central lines inserted (p=<0.001) and who were on TPN (p=<0.001) in the fungal sepsis group compared to the bacterial group (Table 3.12). Table 3.12 Comparison of use of Antibiotics, Central lines, and TPN Between Infants with Fungal and Bacterial Sepsis Fungal Bacterial p-value n=111(%) n=122(%) First line antibiotics Penicillin + Gentamicin 31(28) 60(49) p=0.0018 Antibiotic other than first line 79 (71) 29 (24) <0.001 Tazocin + amikacin 41(37) 13 (11) Meropenem + vancomycin 34 (31) 13 (11) Other 4 (3) 3 (2) Central lines 62 (56) 38 (31) <0.001 Total parenteral nutrition 85 (76) 24 (20) <0.001 40 CHAPTER 4 4. DISCUSSION The advances made in newborn care have resulted in the premature neonate surviving longer and hence prone to interventions and nosocomial infections. Fungal sepsis is emerging as an important cause of sepsis in the neonatal population. One study found an eleven-fold increase in the rate of candidemia.23 Candida species rank as the third and fourth most common organisms isolated in the neonatal units in the UK and in the USA respectively.104 There are few studies on epidemiology of fungal sepsis from developing countries like South Africa. This study was undertaken to determine the epidemiology of Candida infections at the Chris Hani Baragwanath Hospital. The main findings from this study were that 1. the common Candida species causing sepsis in neonates is Candida parapsilosis, 2. the common presenting signs in infants with fungal sepsis are apnoea, respiratory distress and non- specific signs like temperature instability and lethargy, 3. the common clinical diagnosis is necrotizing enterocolitis 4. thrombocytopaenia and a high CRP are found in majority of patients with fungal sepsis, 5. Candida sepsis is associated with high mortality, 6. Thrombocytopaenia and a high CRP are associated with a high mortality, 7. patients with lower birth weight/ gestational age, prolonged hospital stay, low Apgar scores, who have a central venous catheter, TPN and have been treated with Cephalosporins are at risk of having fungal sepsis, 8. In contrast to the common notion that thrombocytopaenia is a sign of fungal sepsis 41 rather than bacterial sepsis, in this study the findings are that thrombocytopaenia is equally common to both bacterial and fungal infected neonates. Studies reported a higher prevalence in the ELBW and VLBW group.24,49 There was also an inverse relationship between gestational age and the acquisition of fungal sepsis. Benjamin et al reported that fungal sepsis is more common if the patient?s gestational age was < 25 weeks.10 In our study the median gestational age was 30 weeks and the birth weight 1500g, which is higher than that reported by Benjamin et al. This difference is accounted for by policies in the unit at the time of the study, where mechanical ventilation was not offered to neonates <1000g. Hence, patients did not live long enough to develop fungal sepsis. The smaller neonates being handled more due to nursing care and blood taking are more prone to infections. This coupled with the immature immune system and thin skin barrier make them more susceptible to nosocomial infections.7 In addition our hospital is a very busy neonatal facility with an average of 3500 admissions per year during the study period. Inadequate hand hygiene techniques could have contributed to the increase in C. parapsilosis in our setting. The literature has identified the hands of health care workers as a source.27,32,35 The median age of onset is 16 days. Saiman et al reported a median age of onset of 14 days for fungal colonisation, and another study documented a mean age of onset for fungal sepsis that ranged from 15 ? 33 days similar findings to our study.7,35 Central venous catheters, TPN and third generation cephalosporins have been identified as risk factors for the development of fungal sepsis in this study. These 42 findings are similar to other studies. Saiman et al identified similar risk factors in addition to the number of days the CVC was left in-situ, the use of H2 ? blockers, and whether the patient was ventilated or not.35 Colonisation was also a risk factor identified as this would increase the likelihood to develop fungal sepsis.62,84,112,113 Factors that Chow et al identified that selected for NAC organisms were the receipt of fluconazole, the insertion of CVCs and the use of antimicrobials.40 Abi Said et al also reported an increase in C. parapsilosis which was associated with the use of the use of TPN and CVCs.22 Candida parapsilosis was the commonest organism isolated. There were no significant differences in the risk factors between the infants who infected with C. parapsilosis compared to the infant infected with C. albicans. Some studies have reported C. albicans as the common organism.35,51,52,80,114 The patients with C. parapsilosis were smaller and had a lower gestational age than the patients with C. albicans. This is similar to the findings by Shoa et al.100 Thrombocytopenia occurs in 20-50% of all neonates admitted to the NICU and the incidence is 50% among preterm infants.115 Thrombocytopenia is used as a non- specific marker for sepsis in the neonate.116 This study showed that patients infected with GN organisms had lower platelet counts than patients with fungal sepsis. Guida et al reported similar findings but used a cut-off of <100000x109/l to define thrombocytopenia.62 We used a platelet count of <150000x109/l in the definition of thrombocytopenia. Bhat et al found a lower platelet count in their GN group of patients.115 They also identified that K.pneumoniae was the organism that 43 caused the most effect on the platelet count. This is due to a lipopolysaccharide component, Lipid A which amplifies the interaction between the IgG Fc receptor and the organism and this causes a further increase in platelet consumption. A similar mechanism has been proposed for fungal sepsis but has not been identified. Another study found no difference in the platelet counts between the GN and gram positive and fungal organisms.117 However, this study did not look at the severity or duration of the thrombocytopenia. These may be important characteristics. Benjamin et al reported Candida organisms to be most associated with thrombocytopenia and formed part of his clinical predictive model.10 The other factors that could cause thrombocytopenia such as maternal steroids, pre- eclampsia, maternal diabetes, maternal ITP were not assessed in this study. All of the C. albicans isolates were sensitive to fluconazole and amphotericin B. The Sentry antimicrobial surveillance programme in the USA reported that 95% of the C. albicans isolates were sensitive to amphotericin B.93 Whilst the TSARY surveillance programme in Taiwan reported an increase in resistance of all isolates to amphotericin B from 0.5% in 1999 to 2.5% in 2002 and a decrease in the resistance rate to fluconazole from 8.45 to 1.9%.79 Yet there were only 3 of the 395 isolates of C. albicans resistant to amphotericin B in TSARY 2002.106 Amongst the C. parapsilosis isolates 93% were susceptible to amphotericin B and 20% were documented to have resistance to fluconazole. This is consistent with the findings in the Artemis Disk antifungal surveillance programme were 20.3% of C. parapsilosis isolates from South Africa were resistant to fluconazole.27 The region with the most susceptible isolates was found in Europe. Clerihew et al reported 44 that all the C. parapsilosis to be sensitive to amphotericin B.36 The C. glabrata isolates were all sensitive to fluconazole and two-thirds showed a susceptible dose-dependent (S-DD) susceptibility profile to amphotericin B. Many studies are reporting an emerging resistance of C. glabrata and C. krusei to both antifungals and the IDSA guidelines recommend a higher dose of these drugs when treating these two isolates.63,102 Fluconazole prophylaxis selects for the isolation of these organisms hence, when treating patients with these two isolates an appropriate antifungal and dose should be commenced. During the period at which the data from this study was collected, fluconazole was not prescribed for prophylaxis. The one C. krusei isolate in this study was sensitive to amphotericin B and had intermediate resistance to fluconazole. This is in contrast to studies where C. krusei is resistant to fluconazole and having decreased susceptibility to amphotericin B.95,118 The significance of these findings is difficult to interpret as there were only two isolates of C. krusei. Fungal sepsis has a mortality rate of 20 ? 40% with the rate being higher in the ELBW (37 ? 40%) as compared to the VLBW infant (32%).47,51 Candida albicans had a higher mortality than C. parapsilosis in this study. This is similar to findings reported by Kauffmann et al where mortality of C. albicans vs C. parapsilosis, was 44% vs 16% respectively. Stoll and Saiman et al had similar findings.8,9,35,109 The higher mortality associated with C. albicans infection is due to the organism?s virulence factors. It tends to adhere tighter to epithelial cells than C. parapsilosis and causes a more lethal clinical presentation.31 Whilst C. parapsilosis is less virulent it is more difficult to eradicate and this compounds the morbidity related to 45 this isolate. Other studies found no difference in mortality between the two organisms.36 Thrombocytopenia is associated with death in patients with Candida sepsis.26,115 It was reported by Bhat and colleagues that mortality was higher; 36% compared to 16% in patients with thrombocytopenia as compared to patients without thrombocytopenia. In this study a significant difference in platelet count was found in the repeat blood count between the survivors and non-survivors. Non-survivors had a much lower platelet count than the survivors. Candida albicans has emerged as the organism with the highest case fatality rate despite being the second commonest organism isolated. Fluconazole prophylaxis in a select group of patients would alter the prevalence of this organism and thereby decrease the overall mortality rate. A randomised control trial would need to be performed in the unit as detailed by Kauffmann as this may alter the epidemiology of fungal sepsis in the neonatal unit. We should then be aware that the prevalence of the NAC may increase and alter the susceptibility profiles of the organisms. The unit policy has subsequently changed where ventilation is offered to neonates < 1000g. This will influence the survival of neonates in our setting and influence the epidemiology of candidemia. The study identified the subset of patients that would develop fungal infections. In our setting if a VLBW had risk factors such as having had central lines, been on TPN and third generation cephalosporins and had abdominal surgery; we would empirically start antifungal treatment. Thrombocytopenia is not a pathognomonic feature as patients with GN sepsis would also present with a low platelet count62. It would be more important to look at the persistence of the thrombocytopenia and to 46 compare the nadir of the low platelet count. Thrombocytopenia is commonly used to identify patients with fungal sepsis but from this study thrombocytopenia was not exclusively associated with fungal sepsis. These patients should be treated with antibacterials first and in cases where patients are not responding antifungal treatment should be considered. At present the choice of antifungals need not be altered as the resistance of the commoner isolates is not posing a serious problem. Trends need to be followed via a surveillance programme so as to identify emerging resistance as is seen in the rest of the world. This would then have an impact on future prescribing practices. 47 CHAPTER 5 CONCLUSION In conclusion the VLBW and smaller infant are more prone to developing fungal sepsis. A lower platelet count is associated with an increased mortality. Having risk factors like the use of TPN, a CVC in-situ and a third generation cephalosporin places the VLBW infant at higher risk for fungal sepsis. Great benefit will be gained if a surveillance programme is undertaken. Fungal sepsis is a major cause of infection and its associated mortality in neonates born or admitted at Chris Hani Baragwanath Hospital. Its epidemiology is similar to what has been reported in other studies. Clinicians need to monitor development of resistance to fluconazole as some Candida species like C. parapsilosis have been found to be resistant to fluconazole. 48 CHAPTER 6 LIMITATIONS This was a retrospective study hence many patient records were not located. The data collection was biased as to what had been written in the patient files. A convenient sample of patients with GN sepsis was used as a comparison to identify risk factors. Hence, they were not matched controls. Furthermore, the GN group was collected over a one year period and the fungal group over a three year period. It would have been ideal to get controls from patients who were not infected. With regard to the laboratory parameters, not all the patients had repeat laboratory markers performed. Had this been a prospective study we would have had increased numbers of repeat specimens and this could alter the trends observed. No other causes for the thrombocytopenia were looked at eg. maternal steroids and preeclampsia or the possibility of a congenital infection. The other pitfall in this study included that we did not look at the duration and nadir of the thrombocytopenia and the change in platelet count from baseline (i.e. before the onset of sepsis). Risk factors that contribute to the development of fungal sepsis are many. We reported on three. We had not taken into account other risk factors such as colonisation rates among the VLBW infants, whether the infant had been ventilated or not, if there was any abdominal surgery and if the neonate had received any steroids or H2 blockers. The number of days that the CVC had been 49 in-situ is another important factor and if the CVC was removed and how this affected clearance rates of the fungal isolate. Susceptibilities for all the organisms were not routinely done by the laboratory at the time, therefore, only two-thirds of the isolates had susceptibilities documented. Hence, the susceptibility profile may be a bit skewed. 50 BIBLIOGRAPHY 1. Neonatal and Perinatal Mortality: Country, Regional and Global Estimates. 2006; Available at: http://www.who.int/making -pregnancy- safer/publications/neonatal.pdf, 2010. 2. The Partnership for Maternal Newborn and Child Health. Opportunities for Africa's newborns: Practical data, policy and programmatic support for neborn care in Africa. : WHO on behalf of the Partnershipfor Maternal Newborn and Child Health; 2006. 3. United Nations Millennium Declaration. 2000; Available at:http://www.unicef.org/mdg/index_aboutthegoals.htm. Accessed April, 2010. 4. Garcia-Prats JA, Cooper TR. The Critically Ill Neonate With Infection:Management Considerations in the Term and Preterm Infant. Seminars in Pediatric infectious Disease 2000;11(1):4-12. 5. Kaufman D, Fairchild KD. Clinical microbiology of bacterial and fungal sepsis in very-low-birth-weight infants. Clin Microbiol Rev 2004 Jul;17(3):638-80, table of contents. 6. Higgins RD, Baker CJ, Raju TN. Executive summary of the workshop on infection in the high-risk infant. J Perinatol 2010 Jun;30(6):379-383. 7. Kaufman D, Fairchild KD. Clinical microbiology of bacterial and fungal sepsis in very-low-birth-weight infants. Clin Microbiol Rev 2004 Jul;17(3):638-80, table of contents. 8. Stoll BJ, Gordon T, Korones SB, Shankaran S, Tyson JE, Bauer CR, et al. Late- onset sepsis in very low birth weight neonates: a report from the National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 1996 Jul;129(1):63-71. 9. Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, Ehrenkranz RA, et al. Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics 2002 Aug;110(2 Pt 1):285-291. 10. Benjamin DK,Jr, DeLong ER, Steinbach WJ, Cotton CM, Walsh TJ, Clark RH. Empirical therapy for neonatal candidemia in very low birth weight infants. Pediatrics 2003 Sep;112(3 Pt 1):543-547. 11. Kapur R, Polin RA, Yoder MC. The Immune System, Developmental Immunology. In: Martin RJ, Fanaroff AA, Walsh MC, editors. Neonatal-Perinatal Medicine. 8th Edition ed. Philedelphia: Mosby, Elsevier; 2006. p. 761. 51 12. Garland JS, Uhing MR. Strategies to prevent bacterial and fungal infection in the neonatal intensive care unit. Clin Perinatol 2009 Mar;36(1):1-13. 13. Butler KM, Baker CJ. Candida: an increasingly important pathogen in the nursery. Pediatr Clin North Am 1988 Jun;35(3):543-563. 14. Pappas PG. Invasive candidiasis. Infect Dis Clin North Am 2006 Sep;20(3):485-506. 15. Eggimann P, Garbino J, Pittet D. Epidemiology of Candida species infections in critically ill non-immunosuppressed patients. Lancet Infect Dis 2003 Nov;3(11):685-702. 16. Hazen KC, Howell SA. Candida, Cryptococcus and other Yeasts of Medical Importance. In: Murray PR, Baron EJ, Jorgensen JH, Laundry ML, Pfaller MA, editors. Manual of Clinical Microbiology. 9th Edition ed. Washington, D.C.: ASM PRESS; 2007. p. 1764-1780. 17. Edwards MS. Fungal And Protozoal Infections. In: Martin RJ, Fanaroff AA, Walsh MC, editors. Neonatal-Perinatal Medicine. 8th Edition ed. Philedelphia: Mosby, Elsevier; 2006. p. 830. 18. Edwards JEJ. Mycoses, Candida Species. In: Mandell GL, Bennett JE, Dolin R, editors. Principles nad Practices of Infectious Diseases. 6th Edition ed. Philedelphia: Elsevier, Chrchill, Livingstone; 2005. p. 2938-2951. 19. Benjamin DK,Jr, Garges H, Steinbach WJ. Candida bloodstream infection in neonates. Semin Perinatol 2003 Oct;27(5):375-383. 20. Chen SC, Marriott D, Playford EG, Nguyen Q, Ellis D, Meyer W, et al. Candidaemia with uncommon Candida species: predisposing factors, outcome, antifungal susceptibility, and implications for management. Clin Microbiol Infect 2009 Jul;15(7):662-669. 21. Nguyen MH, Peacock JE,Jr, Morris AJ, Tanner DC, Nguyen ML, Snydman DR, et al. The changing face of candidemia: emergence of non-Candida albicans species and antifungal resistance. Am J Med 1996 Jun;100(6):617-623. 22. Abi-Said D, Anaissie E, Uzun O, Raad I, Pinzcowski H, Vartivarian S. The epidemiology of hematogenous candidiasis caused by different Candida species. Clin Infect Dis 1997 Jun;24(6):1122-1128. 23. Kossoff EH, Buescher ES, Karlowicz MG. Candidemia in a neonatal intensive care unit: trends during fifteen years and clinical features of 111 cases. Pediatr Infect Dis J 1998 Jun;17(6):504-508. 52 24. Presterl E, Daxbock F, Graninger W, Willinger B. Changing pattern of candidaemia 2001-2006 and use of antifungal therapy at the University Hospital of Vienna, Austria. Clin Microbiol Infect 2007 Nov;13(11):1072-1076. 25. Pfaller MA, Jones RN, Doern GV, Sader HS, Hollis RJ, Messer SA. International surveillance of bloodstream infections due to Candida species: frequency of occurrence and antifungal susceptibilities of isolates collected in 1997 in the United States, Canada, and South America for the SENTRY Program. The SENTRY Participant Group. J Clin Microbiol 1998 Jul;36(7):1886-1889. 26. Chen TC, Chen YH, Tsai JJ, Peng CF, Lu PL, Chang K, et al. Epidemiologic analysis and antifungal susceptibility of Candida blood isolates in southern Taiwan. J Microbiol Immunol Infect 2005 Jun;38(3):200-210. 27. Pfaller MA, Diekema DJ, Gibbs DL, Newell VA, Ng KP, Colombo A, et al. Geographic and temporal trends in isolation and antifungal susceptibility of Candida parapsilosis: a global assessment from the ARTEMIS DISK Antifungal Surveillance Program, 2001 to 2005. J Clin Microbiol 2008 Mar;46(3):842-849. 28. Hobson RP. The global epidemiology of invasive Candida infections--is the tide turning? J Hosp Infect 2003 Nov;55(3):159-68; quiz 233. 29. Odds FC. Candida infections: an overview. Crit Rev Microbiol 1987;15(1):1-5. 30. Leibovitz E. Neonatal candidosis: clinical picture, management controversies and consensus, and new therapeutic options. J Antimicrob Chemother 2002 Feb;49 Suppl 1:69-73. 31. Bendel CM. Colonization and epithelial adhesion in the pathogenesis of neonatal candidiasis. Semin Perinatol 2003 Oct;27(5):357-364. 32. van Asbeck EC, Huang YC, Markham AN, Clemons KV, Stevens DA. Candida parapsilosis fungemia in neonates: genotyping results suggest healthcare workers hands as source, and review of published studies. Mycopathologia 2007 Dec;164(6):287-293. 33. Weems JJ,Jr. Candida parapsilosis: epidemiology, pathogenicity, clinical manifestations, and antimicrobial susceptibility. Clin Infect Dis 1992 Mar;14(3):756-766. 34. Trofa D, Gacser A, Nosanchuk JD. Candida parapsilosis, an emerging fungal pathogen. Clin Microbiol Rev 2008 Oct;21(4):606-625. 35. Saiman L, Ludington E, Dawson JD, Patterson JE, Rangel-Frausto S, Wiblin RT, et al. Risk factors for Candida species colonization of neonatal intensive care unit patients. Pediatr Infect Dis J 2001 Dec;20(12):1119-1124. 53 36. Clerihew L, Lamagni TL, Brocklehurst P, McGuire W. Candida parapsilosis infection in very low birthweight infants. Arch Dis Child Fetal Neonatal Ed 2007 Mar;92(2):F127-9. 37. Rowen JL. Fungal Infections in the Neonatal Intensive Care Unit. Seminars in Pediatric Infectious Diseases 2001;12(2):107-114. 38. Cheng MF, Yang YL, Yao TJ, Lin CY, Liu JS, Tang RB, et al. Risk factors for fatal candidemia caused by Candida albicans and non-albicans Candida species. BMC Infect Dis 2005 Apr 7;5(1):22. 39. Cheng MF, Yang YL, Yao TJ, Lin CY, Liu JS, Tang RB, et al. Risk factors for fatal candidemia caused by Candida albicans and non-albicans Candida species. BMC Infect Dis 2005 Apr 7;5(1):22. 40. Chow JK, Yolan Y, Ruthazer R. Risk factors for albicans and non-albicans candidemia in the intensive care unit. Crit Care Med 2008;36(7):1993-1998. 41. Samaranayake YH, Samaranayake LP. Candida krusei: biology, epidemiology, pathogenicity and clinical manifestations of an emerging pathogen. J Med Microbiol 1994 Nov;41(5):295-310. 42. Pfaller MA, Diekema DJ, Gibbs DL, Newell VA, Nagy E, Dobiasova S, et al. Candida krusei, a multidrug-resistant opportunistic fungal pathogen: geographic and temporal trends from the ARTEMIS DISK Antifungal Surveillance Program, 2001 to 2005. J Clin Microbiol 2008 Feb;46(2):515-521. 43. Mean M, Marchetti O, Calandra T. Bench-to-bedside review: Candida infections in the intensive care unit. Crit Care 2008;12(1):204. 44. Fidel PL,Jr, Vazquez JA, Sobel JD. Candida glabrata: review of epidemiology, pathogenesis, and clinical disease with comparison to C. albicans. Clin Microbiol Rev 1999 Jan;12(1):80-96. 45. Askin DF. Bacterial and fungal infections in the neonate. J Obstet Gynecol Neonatal Nurs 1995 Sep;24(7):635-643. 46. Benjamin DK,Jr, Garges H, Steinbach WJ. Candida bloodstream infection in neonates. Semin Perinatol 2003 Oct;27(5):375-383. 47. Kaufman D. Fungal infection in the very low birthweight infant. Curr Opin Infect Dis 2004 Jun;17(3):253-259. 48. Chapman RL. Candida infections in the neonate. Curr Opin Pediatr 2003 Feb;15(1):97-102. 54 49. Chapman RL, Faix RG. Invasive neonatal candidiasis: an overview. Semin Perinatol 2003 Oct;27(5):352-356. 50. Chapman RL. Prevention and treatment of Candida infections in neonates. Semin Perinatol 2007 Feb;31(1):39-46. 51. Shetty SS, Harrison LH, Hajjeh RA, Taylor T, Mirza SA, Schmidt AB, et al. Determining risk factors for candidemia among newborn infants from population- based surveillance: Baltimore, Maryland, 1998-2000. Pediatr Infect Dis J 2005 Jul;24(7):601-604. 52. Feja KN, Wu F, Roberts K, Loughrey M, Nesin M, Larson E, et al. Risk factors for candidemia in critically ill infants: a matched case-control study. J Pediatr 2005 Aug;147(2):156-161. 53. Brecht M, Clerihew L, McGuire W. Prevention and treatment of invasive fungal infection in very low birthweight infants. Arch Dis Child Fetal Neonatal Ed 2009 Jan;94(1):F65-9. 54. Baley JE, Kliegman RM, Boxerbaum B, Fanaroff AA. Fungal colonization in the very low birth weight infant. Pediatrics 1986 Aug;78(2):225-232. 55. Huang YC, Li CC, Lin TY, Lien RI, Chou YH, Wu JL, et al. Association of fungal colonization and invasive disease in very low birth weight infants. Pediatr Infect Dis J 1998 Sep;17(9):819-822. 56. Magny JF, Bremard-Oury C, Brault D, Menguy C, Voyer M, Landais P, et al. Intravenous immunoglobulin therapy for prevention of infection in high-risk premature infants: report of a multicenter, double-blind study. Pediatrics 1991 Sep;88(3):437-443. 57. Benjamin DK,Jr, Ross K, McKinney RE,Jr, Benjamin DK, Auten R, Fisher RG. When to suspect fungal infection in neonates: A clinical comparison of Candida albicans and Candida parapsilosis fungemia with coagulase-negative staphylococcal bacteremia. Pediatrics 2000 Oct;106(4):712-718. 58. Cotten CM, McDonald S, Stoll B, Goldberg RN, Poole K, Benjamin DK,Jr, et al. The association of third-generation cephalosporin use and invasive candidiasis in extremely low birth-weight infants. Pediatrics 2006 Aug;118(2):717-722. 59. Karlowicz MG, Hashimoto LN, Kelly RE,Jr, Buescher ES. Should central venous catheters be removed as soon as candidemia is detected in neonates? Pediatrics 2000 Nov;106(5):E63. 60. Linder N, Levit O, Klinger G, Kogan I, Levy I, Shalit I, et al. Risk factors associated with candidaemia in the neonatal intensive care unit: a case-control study. J Hosp Infect 2004 Aug;57(4):321-324. 55 61. Benjamin DK,Jr, Stoll BJ, Fanaroff AA, McDonald SA, Oh W, Higgins RD, et al. Neonatal candidiasis among extremely low birth weight infants: risk factors, mortality rates, and neurodevelopmental outcomes at 18 to 22 months. Pediatrics 2006 Jan;117(1):84-92. 62. Guida JD, Kunig AM, Leef KH, McKenzie SE, Paul DA. Platelet count and sepsis in very low birth weight neonates: is there an organism-specific response? Pediatrics 2003 Jun;111(6 Pt 1):1411-1415. 63. Pappas PG, Kauffman CA, Andes D, Benjamin DK,Jr, Calandra TF, Edwards JE,Jr, et al. Clinical practice guidelines for the management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis 2009 Mar 1;48(5):503-535. 64. Chapman SW, Sullivan DC, Cleary JD. In search of the holy grail of antifungal therapy. Trans Am Clin Climatol Assoc 2008;119:197-215; discussion 215-6. 65. Mukherjee PK, Sheehan DJ, Hitchcock CA, Ghannoum MA. Combination treatment of invasive fungal infections. Clin Microbiol Rev 2005 Jan;18(1):163-194. 66. Pappas PG, Rex JH, Sobel JD, Filler SG, Dismukes WE, Walsh TJ, et al. Guidelines for treatment of candidiasis. Clin Infect Dis 2004 Jan 15;38(2):161-189. 67. Masia Canuto M, Gutierrez Rodero F. Antifungal drug resistance to azoles and polyenes. Lancet Infect Dis 2002 Sep;2(9):550-563. 68. Dupont B. Overview of the lipid formulations of amphotericin B. J Antimicrob Chemother 2002 Feb;49 Suppl 1:31-36. 69. Adler-Moore JP, Proffitt RT. Amphotericin B lipid preparations: what are the differences? Clin Microbiol Infect 2008 May;14 Suppl 4:25-36. 70. Juster-Reicher A, Leibovitz E, Linder N, Amitay M, Flidel-Rimon O, Even-Tov S, et al. Liposomal amphotericin B (AmBisome) in the treatment of neonatal candidiasis in very low birth weight infants. Infection 2000 Jul-Aug;28(4):223-226. 71. Adler-Moore J, Proffitt RT. AmBisome: liposomal formulation, structure, mechanism of action and pre-clinical experience. J Antimicrob Chemother 2002 Feb;49 Suppl 1:21-30. 72. Juster-Reicher A, Flidel-Rimon O, Amitay M, Even-Tov S, Shinwell E, Leibovitz E. High-dose liposomal amphotericin B in the therapy of systemic candidiasis in neonates. Eur J Clin Microbiol Infect Dis 2003 Oct;22(10):603-607. 73. Linder N, Klinger G, Shalit I, Levy I, Ashkenazi S, Haski G, et al. Treatment of candidaemia in premature infants: comparison of three amphotericin B preparations. J Antimicrob Chemother 2003 Oct;52(4):663-667. 56 74. Moreira ME. Controversies about the management of invasive fungal infections in very low birth weight infants. J Pediatr (Rio J) 2005 Mar;81(1 Suppl):S52-8. 75. Karlowicz MG. Candidal renal and urinary tract infection in neonates. Semin Perinatol 2003 Oct;27(5):393-400. 76. Driessen M, Ellis JB, Cooper PA, Wainer S, Muwazi F, Hahn D, et al. Fluconazole vs. amphotericin B for the treatment of neonatal fungal septicemia: a prospective randomized trial. Pediatr Infect Dis J 1996 Dec;15(12):1107-1112. 77. Driessen M, Ellis JB, Muwazi F, De Villiers FP. The treatment of systemic candidiasis in neonates with oral fluconazole. Ann Trop Paediatr 1997 Sep;17(3):263-271. 78. De Rosa FG, Garazzino S, Pasero D, Di Perri G, Ranieri VM. Invasive candidiasis and candidemia: new guidelines. Minerva Anestesiol 2009 Jul- Aug;75(7-8):453-458. 79. Yang YL, Li SY, Cheng HH, Lo HJ, TSARY Hospitals. The trend of susceptibilities to amphotericin B and fluconazole of Candida species from 1999 to 2002 in Taiwan. BMC Infect Dis 2005 Nov 3;5:99. 80. Nakamura T, Takahashi H. Epidemiological study of Candida infections in blood: susceptibilities of Candida spp. to antifungal agents, and clinical features associated with the candidemia. J Infect Chemother 2006 Jun;12(3):132-138. 81. Kaufman D, Boyle R, Hazen KC, Patrie JT, Robinson M, Donowitz LG. Fluconazole prophylaxis against fungal colonization and infection in preterm infants. N Engl J Med 2001 Dec 6;345(23):1660-1666. 82. Kaufman D. Strategies for prevention of neonatal invasive candidiasis. Semin Perinatol 2003 Oct;27(5):414-424. 83. Kaufman D, Boyle R, Hazen KC, Patrie JT, Robinson M, Grossman LB. Twice weekly fluconazole prophylaxis for prevention of invasive Candida infection in high-risk infants of <1000 grams birth weight. J Pediatr 2005 Aug;147(2):172-179. 84. Manzoni P, Arisio R, Mostert M, Leonessa M, Farina D, Latino MA, et al. Prophylactic fluconazole is effective in preventing fungal colonization and fungal systemic infections in preterm neonates: a single-center, 6-year, retrospective cohort study. Pediatrics 2006 Jan;117(1):e22-32. 85. Montravers P, Jabbour K. Clinical consequences of resistant Candida infections in intensive care. Int J Antimicrob Agents 2006 Jan;27(1):1-6. 57 86. Sarvikivi E, Lyytikainen O, Soll DR, Pujol C, Pfaller MA, Richardson M, et al. Emergence of fluconazole resistance in a Candida parapsilosis strain that caused infections in a neonatal intensive care unit. J Clin Microbiol 2005 Jun;43(6):2729- 2735. 87. Clerihew L, Austin N, McGuire W. Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants. Cochrane Database Syst Rev 2007 Oct 17;(4)(4):CD003850. 88. Clerihew L, Austin N, McGuire W. Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants. Cochrane Database Syst Rev 2007 Oct 17;(4)(4):CD003850. 89. Swoboda SM, Merz WG, Lipsetta PA. Candidemia: the impact of antifungal prophylaxis in a surgical intensive care unit. Surg Infect (Larchmt) 2003 Winter;4(4):345-354. 90. Baddley J W, Pappas P G. Antifungal Combination Therapy. Drugs 2005;65(11):1461-1480. (91) Rex JH, Walsh TJ, Sobel JD, Filler SG, Pappas PG, Dismukes WE, et al. Practice guidelines for the treatment of candidiasis. Infectious Diseases Society of America. Clin Infect Dis 2000 Apr;30(4):662-678. 92. Sanglard D, Odds FC. Resistance of Candida species to antifungal agents: molecular mechanisms and clinical consequences. Lancet Infect Dis 2002 Feb;2(2):73-85. 93 Walsh TJ, Lee J, Lecciones J, Rubin M, Butler K, Francis P, et al. Empiric therapy with amphotericin B in febrile granulocytopenic patients. Rev Infect Dis 1991 May-Jun;13(3):496-503. 94. Marr KA. Empirical antifungal therapy--new options, new tradeoffs. N Engl J Med 2002 Jan 24;346(4):278-280. 95. Pfaller MA, Diekema DJ, Jones RN, Sader HS, Fluit AC, Hollis RJ, et al. International surveillance of bloodstream infections due to Candida species: frequency of occurrence and in vitro susceptibilities to fluconazole, ravuconazole, and voriconazole of isolates collected from 1997 through 1999 in the SENTRY antimicrobial surveillance program. J Clin Microbiol 2001 Sep;39(9):3254-3259. 96. Rex JH, Pfaller MA, Walsh TJ, Chaturvedi V, Espinel-Ingroff A, Ghannoum MA, et al. Antifungal susceptibility testing: practical aspects and current challenges. Clin Microbiol Rev 2001 Oct;14(4):643-58, table of contents. 97. Rex JH, Pfaller MA, Rinaldi MG, Polak A, Galgiani JN. Antifungal susceptibility testing. Clin Microbiol Rev 1993 Oct;6(4):367-381. 58 98. Pfaller MA, Rex JH, Rinaldi MG. Antifungal susceptibility testing: technical advances and potential clinical applications. Clin Infect Dis 1997 May;24(5):776- 784. 99. Ellis D. Amphotericin B: spectrum and resistance. J Antimicrob Chemother 2002 Feb;49 Suppl 1:7-10. 100. Shao PL, Huang LM, Hsueh PR. Recent advances and challenges in the treatment of invasive fungal infections. Int J Antimicrob Agents 2007 Dec;30(6):487-495. 101. Almirante B, Rodriguez D, Park BJ, Cuenca-Estrella M, Planes AM, Almela M, et al. Epidemiology and predictors of mortality in cases of Candida bloodstream infection: results from population-based surveillance, barcelona, Spain, from 2002 to 2003. J Clin Microbiol 2005 Apr;43(4):1829-1835. 102. Cheng MF, Yu KW, Tang RB, Fan YH, Yang YL, Hsieh KS, et al. Distribution and antifungal susceptibility of Candida species causing candidemia from 1996 to 1999. Diagn Microbiol Infect Dis 2004 Jan;48(1):33-37. 103. Hajjeh RA, Sofair AN, Harrison LH, Lyon GM, Arthington-Skaggs BA, Mirza SA, et al. Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program. J Clin Microbiol 2004 Apr;42(4):1519-1527. 104. Leroy O, Gangneux JP, Montravers P, Mira JP, Gouin F, Sollet JP, et al. Epidemiology, management, and risk factors for death of invasive Candida infections in critical care: a multicenter, prospective, observational study in France (2005-2006). Crit Care Med 2009 May;37(5):1612-1618. 105. Price MF, LaRocco MT, Gentry LO. Fluconazole susceptibilities of Candida species and distribution of species recovered from blood cultures over a 5-year period. Antimicrob Agents Chemother 1994 Jun;38(6):1422-1424. 106. Yang YL, Li SY, Cheng HH, Lo HJ, TSARY Hospitals. Susceptibilities to amphotericin B and fluconazole of Candida species in TSARY 2002. Diagn Microbiol Infect Dis 2005 Mar;51(3):179-183. 107. Krcmery V, Barnes AJ. Non-albicans Candida spp. causing fungaemia: pathogenicity and antifungal resistance. J Hosp Infect 2002 Apr;50(4):243-260. 108. Faix RG, Chapman RL. Central nervous system candidiasis in the high-risk neonate. Semin Perinatol 2003 Oct;27(5):384-392. 109. Stoll BJ, Hansen N. Infections in VLBW infants: studies from the NICHD Neonatal Research Network. Semin Perinatol 2003 Aug;27(4):293-301. 59 110. Gudlaugsson O, Gillespie S, Lee K, Vande Berg J, Hu J, Messer S, et al. Attributable mortality of nosocomial candidemia, revisited. Clin Infect Dis 2003 Nov 1;37(9):1172-1177. 111. Pappas PG, Rex JH, Lee J, Hamill RJ, Larsen RA, Powderly W, et al. A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis 2003 Sep 1;37(5):634-643. 112. Manzoni P, Farina D, Leonessa M, d'Oulx EA, Galletto P, Mostert M, et al. Risk factors for progression to invasive fungal infection in preterm neonates with fungal colonization. Pediatrics 2006 Dec;118(6):2359-2364. 113. Kaufman DA, Gurka MJ, Hazen KC, Boyle R, Robinson M, Grossman LB. Patterns of fungal colonization in preterm infants weighing less than 1000 grams at birth. Pediatr Infect Dis J 2006 Aug;25(8):733-737. 114. Huang YC, Lin TY, Lien RI, Chou YH, Kuo CY, Yang PH, et al. Candidaemia in special care nurseries: comparison of albicans and parapsilosis infection. J Infect 2000 Mar;40(2):171-175. 115. Bhat MA, Bhat JI, Kawoosa MS, Ahmad SM, Ali SW. Organism-specific platelet response and factors affecting survival in thrombocytopenic very low birth weight babies with sepsis. J Perinatol 2009 Oct;29(10):702-708. 116. Akarsu S, Taskin E, Kilic M, Ozdiller S, Gurgoze MK, Yilmaz E, et al. The effects of different infectious organisms on platelet counts and platelet indices in neonates with sepsis: is there an organism-specific response? J Trop Pediatr 2005 Dec;51(6):388-391. 117. Manzoni P, Mostert M, Galletto P, Gastaldo L, Gallo E, Agriesti G, et al. Is thrombocytopenia suggestive of organism-specific response in neonatal sepsis? Pediatr Int 2009 Apr;51(2):206-210. 118. Pfaller MA, Diekema DJ, Jones RN, Messer SA, Hollis RJ, SENTRY Participants Group. Trends in antifungal susceptibility of Candida spp. isolated from pediatric and adult patients with bloodstream infections: SENTRY Antimicrobial Surveillance Program, 1997 to 2000. J Clin Microbiol 2002 Mar;40(3):852-856. 60 APPENDIX A 61 62 63 64 65 66 67 68 69 70 71 DATA COLLECTION SHEET: FUNGAL SEPSIS 1. Infant details ID Number: ????????????????????.. Culture date: ......./??./??. ; Episode (No): ??. ; Ward: ???. ; Admission date: ??./??./??. Patient?s Hosp No: ????????????????????? ; Date of Birth: ??./??./??. Place of Birth: ???? (1=Bara, 2=Clinic, 3=Home, 4=Other Hosp) Birthwt: ???. ; GA: ???. ; (Based o Obstets/Paeds): ??? ; Sex: ??.. ; Race: ??????.. ; Apgar: 1??../5???.. Mode of delivery: C/S, NVD, Forceps, Vacuum Maternal HIV status: Pos / Neg / Unknown 2. Presentation Date of onset of symptoms: ??./??./??. ; Day of life at onset: ???????????.. Symptoms (Reason for sepsis work up): ?????????????????????????????????????????????.. Diagnosis: ?????????????????????????????????? Antibiotics started on (date): ??????????????????????.. Previous antibiot. Stopped: ??./??./??. Date previous antibiot. Stopped: ??./??./??. Ventilated: Yes / No; If Yes, Date intubated: ??./??./??. ; Date extubated: ??./??./??. On Oxygen Headbox / Cannulae: Yes / No Central line: Yes / No; Type: UAC/UVC/Other; Date inserted: ??./??./??. ; Date removed: ??./??./??. Peripheral line: Yes / No TPN: Yes / No, If yes, Date started, ??./??./??. ? Intralipids: Yes / No APPENDIX B 72 FBC done with sepsis w/u: Wbc: ??. Hb: ??. Plts: ??. (Date of FBC: ??./??./??. ) Diff Count: Neutr??.; Lymph: ??..; Mono: ??.; Eos: ??; Baso: ??. Others: ?????.. Smear results: ??????????????????????????????????????????????????????? CPR done with sepsis w/u: ???. CSF done with or around time of sepsis w/u: Yes / No; If Yes , date CSF done: ??./??./??. CSF Results: Polys: ????; Lymphs: ???.; Rbcs: ???.; Protein: ???.; Gluc: ???.; CI: ???. FBC immed. After sepsis w/u: Wbc: ???. Hb: ???. Plts: ???. (Date of FBC: ??../??../??.. Diff Count: Neutr ???.; Lymph:???.; Mono: ???.; Eos: ???. Baso:???. Others: ???. Smear results: ??????????????????????????????????????????????????????? Repeat CRP done: ???????. CSF Repeated: Yes / No; If Yes, date CSF repeated: ???/???./???. CSF Results: Polys: ???.; Lymphs: ???.; Rbcs: ???.; Protein: ???.; Gluc:???.; CI:???. 3. Organism and susceptibilities Organism 1: ????????????????????????????.; Culture Site: Bld / CSF If fungal, Sensitivity: Fluconazole ? R/S; Amphotericin B-R/S Organism 2: ????????????????????????????.; Culture Site: Bld / CSF If fungal, Sensitivity: Fluconazole- R/S; Amphotericin B- R/S Culture repeated: Yes / No; Date culture repeated: ???../???/???; Repeat culture: Positive / Negative If culture positive, Organism: ??????????????????..; If fungal, Sensitivity: Fluconazole ? R/S; Amphotericin B- R/S 73 4. Outcome Died: Yes / No; If died, date of death: ???/???./???. Cause of death: ???????????????????????????????????????????????. Postmortem: Yes / No; If done, diagnosis after postmortem:????????????????????.. Other Comments:??????????????????????????????????????????????. 74 APPENDIX C 75