The copyright of this thesis vests in the author. No quotation from it or information derived from it is to be published without full acknowledgementTown of the source. The thesis is to be used for private study or non- commercial research purposes only. Cape Published by the University ofof Cape Town (UCT) in terms of the non-exclusive license granted to UCT by the author. University A Study of Determinants and Prevalence of Rheumatic Heart Disease in Cape Town Mark Emmanuel Engel Town Thesis Presented for the Degree of Cape DOCTOR OF ofPHILOSOPHY In the Department of Medicine UNIVERSITY OF CAPE TOWN University Supervisor: Prof Bongani Mayosi October 2012 PhD Thesis: Mark E Engel A STUDY OF DETERMINANTS AND PREVALENCE OF RHEUMATIC HEART DISEASE IN CAPE TOWN Town Cape of University PhD Thesis: Mark E Engel ad maiorem Dei gloriam Town Cape of University PhD Thesis: Mark E Engel | Page iii Publications arising from this thesis • DA Watkins, LJ Zühlke, ME Engel, BM Mayosi. Rheumatic Fever: Neglected Again. Science 2009; 324:37 (Letter) • ME Engel, L Zühlke, K Robertson. A.S.A.P. Programme in Rheumatic Fever and Rheumatic Heart Disease: Where are we now in South Africa? SA Heart 2009; 6:270-3 (Commentary) • Zühlke L, Engel ME. Rheumatic Heart Disease: The time has come. Clinical Epidemiology 2011; 3: 171. (Letter) • Engel, ME; Stander, R; Vogel, J; Adeyemo, AA; Mayosi, BM. Genetic susceptibility to acute rheumatic fever: a systematic review and meta-analysis of twin studies. PLoS One. 2011;6(9):e25326. Town Cape of University PhD Thesis: Mark E Engel | Page iv Acknowledgements • This thesis was supported in part by funds from the Life Foundation, World Heart Federation, National Institutes of Health, USA, Medical Research Council, Else Kröner Frasenius Foundation Wellcome Trust/CIDRI, and University of Cape Town. • Professor Bongani Mayosi (supervisor and principal investigator): This thesis would not have been possible without his support, excellent guidance and enthusiastic encouragement. • Dr Liesl Zühlke: her guidance, support and recommendations to this thesis were invaluable. • ASAP team members, with special mention of Mr Simpiwe Nkepu, Ms Carolise Lemmer:, Sisters Veronica and Sophia and Mr Dylan Barth. Their tenacity at recruiting patients and providing high quality fieldwork data is the backbone of this thesis. • Dr Jim Dale for his vision, guidance and continued support throughout the studies. • Dr Debo Adeyemo: His guidance and statistical support were aTown noteworthy contribution to Study 1. • Mr Babu Mohamed: His assistance with genotyping of the samples is greatly appreciated. • Members of the clinic-based study team with specialCape mention of Dr Andrew Whitelaw, Prof Gary Maartens and Dr Ronald Gounden for their guidance and clinical input. of • The staff of the Department of Medicine, including administrative staff, students and friends for creating such a pleasant working environment. • Colleagues who have given advice along the way with special mention of Martie Muller and Motasim Badri for their statistical guidance. • Julia and Jana for assistance with the translation of Russian articles included in the systematic review • Last and by noUniversity means least, I acknowledge the love, understanding, prayers and relentless support of my parents, wife Janet, son Hayden and daughter, Ms Emily. PhD Thesis: Mark E Engel | Page v Abbreviations A.S.A.P. Awareness, Surveillance, Advocacy, Prevention Programme in ARF/RHD ARF Acute rheumatic fever β beta βHS beta-Haemolytic streptococcus CDC Centres for Disease Control and Prevention CI Confidence interval CPRs Clinical prediction rules CRF Case report form DZ Dizygotic Town GAS Group A streptococcus MZ Monozygotic OR Odds ratio Cape PPS Probability proportional to size of RHD Rheumatic heart disease ROC Receiver operating curve SA South Africa WHF World UniversityHeart Federation WHO World Health Organisation PhD Thesis: Mark E Engel | Page vi List of Figures Figure 1.1 Progression from GAS infection to established RHD 2 Figure 1.2 Electron micrograph of group A M type 24 streptococci 3 Figure 1.3 Pattern of haemolysis around streptococcal colonies 5 Figure 1.4 Schematic illustration of M protein 6 Figure 1.5 Percentage killing by bactericidal antibodies evoked by the 30-valent vaccine. 8 Figure 1.6 Pathogenic pathway for ARF and RHD 9 Figure 1.7 Major events triggering rheumatic valvular lesions in RHD 10 Figure 1.8 Attachment and internalization of streptococci by human culturedTown pharyngeal cells 15 Figure 1.9 Light microscopy image of mitral valve from RHD patients 27 Figure 1.10 Rheumatic heart disease prevalence rates in children 28 Figure 1.11 Declining rates of death due to Rheumatic Fever.Cape 31 Figure 1.12 Trends of RHD prevalence per 1000 ofpersons for each WHO region, 32 Figure 1.13 Prevalence of GAS Infection Among Children Presenting With Sore Throat 37 Figure 1.14 Pooled prevalence rates of echocardiographically-confirmed WHO-defined Definite 42 or Probable RHD in school children. Figure 2.1 The Vanguard Community Demonstration Site 53 Figure 2.2 The customisedUniversity clinic used for screening school children 54 Figure 2.3 Primary and high schools within Langa and Bonteheuwel 68 Figure 2.4 Echocardiographic criteria used in the Vanguard Screening study 74 Figure 3.1 Process of arriving at final set of included studies 83 Figure 3.2 Odds ratio of concordance for acute rheumatic fever according to type of zygosity 88 Figure 4.1 Age distribution of GAS in asymptomatic school children 99 Figure 4.2 Distribution of emm types in GAS isolated from asymptomatic school children 101 PhD Thesis: Mark E Engel | Page vii List of figures, cont’ Figure 5.1 Gender-specific GAS isolation rates by community within Vanguard Demonstration 113 Site Figure 5.2 Seasonal variation in the proportion of culture-positive GAS isolated in the 114 Vanguard Demonstration Site within each year Figure 5.3 Estimated incidence of GAS culture-positive pharyngitis in Bonteheuwel and Langa 116 children Figure 7.1 Age and Gender Distribution of participants from the Bonteheuwel and Langa 141 School Districts, Cape Town, South Africa Figure 7.2 Prevalence of echocardiographically-confirmed RHD based on WHO or WHO 146 modified criteria for Definite and Probable RHD in school children aged < 18 years. Town Cape of University PhD Thesis: Mark E Engel | Page viii List of Tables Table 1.1 Summary of clinical prediction rules 21 Table 1.2 GAS carriage among asymptomatic children in countries outside of South Africa 35 Table 1.3 Incidence of symptomatic GAS pharyngitis among children 38 Table 1.4 Prevalence of RHD among asymptomatic school children 42 Table 2.1 Vanguard Demonstration Site: Demographic characteristics with respect to 55 suburb. Table 2.2 Age distribution amongst Vanguard children 56 Table 2.3 Expected sample size distribution by region 76 Table 3.1 Studies that were excluded from the systematic review Town 84 Table 3.2 Observational studies of zygosity and concordance for rheumatic fever 86 Table 3.3 Possible sources of bias: defining characteristics in individual studies 87 Table 3.4 Results of fitting the ‘ACE’ threshold model toCape estimate genetic and environmental 90 components of variance of Table 4.1 Comparison of some of the characteristics with respect to GAS status, of the 98 asymptomatic school children recruited into the study during February 2009 – November 2011 Table 4.2 Seasonal distribution of beta haemolytic streptococci isolates from asymptomatic 100 school children Table 5.1 Demographic Data for Enrolled Participants According to Clinic 110 Table 5.2 ComparisonUniversity of GAS-Positive Throat Cultures According to Community 112 Table 5.3 Comparison of GAS-Positive Throat Cultures According to Study Year 115 Table 5.4 Estimated Incidence of Sore Throat and Primary Cases of GAS Culture-Positive 117 Pharyngitis Table 6.1 Comparison of the characteristics of patients presenting with pharyngitis in the 127 derivation and validation sets PhD Thesis: Mark E Engel | Page ix List of tables, cont’ Table 6.2 Performance of 3 clinical decision rules for streptococcal pharyngitis in children 129 aged 3-15 years comprising the derivation set (n=666) from the Vanguard Demonstration Site Table 6.3 Results of Univariable Analysis of Potential Predictive Variables for GAS in 130 children aged 3-15 years with pharyngitis Table 6.4 Multivariable Analysis of Potential Predictive Variables for GAS in children aged 5- 131 15 years with pharyngitis Table 6.5 Scorecard based on multivariable analysis model to calculate the probability of 132 GAS positivity in children aged 3 - 15 years with pharyngitis Table 6.6 Observed percentage of GAS-positive patients within a specific range of scores 132 Table 6.7 Performance of the Cape Town Clinical Prediction Rule for streptococcal 133 pharyngitis using a cut-off score of ≥ 3 in children aged 3 - Town15 years in the Vanguard population validation set. Table 7.1 Characteristics of the participants included in the school-based RHD screening 142 study Cape Table 7.2 Prevalence rates of echocardiography-confirmed definite and probable RHD 143 among asymptomatic school childrenof University PhD Thesis: Mark E Engel | Page x Abstract BACKGROUND Rheumatic Heart Disease (RHD) is a post-infectious immune disease ascribed to an interaction between a rheumatogenic strain of group A streptococcus, a susceptible host who lives in poor social conditions with limited access to medical facilities. The disease process begins with repeated group A streptococcal (GAS) infections, which, subsequently result in acute rheumatic fever (ARF). In the absence of intervention, repeated bouts of ARF in turn, may result in progression to RHD, particularly in those ARF patients with cardiac involvement. The prevalence of ARF and RHD in developedTown countries has shown considerable decline during the last century, largely attributed to improved living conditions and access to healthcare. Epidemiological data from developing countries, Cape while scant, indicate a continued high prevalence of GAS-positive pharyngitis and RHD.