Economic studies of chronic kidney disease

Patients with end-stage renal disease ESRD, the most advanced form of CKD, comprise 0.15% of the global population but consume 2-4% healthcare budgets.. 48 Chapter 2 HEALTH-RELATED QUALI

ECONOMIC STUDIES OF CHRONIC KIDNEY DISEASE

A thesis submitted for the Degree of

4 The composition of the thesis is my own work

ii

ABSTRACT

Approximately 9% of the global population has chronic kidney disease (CKD) Patients with end-stage renal disease (ESRD), the most advanced form of CKD, comprise 0.15% of the global population but consume 2-4% healthcare budgets While the impact of CKD and ESRD are experienced most immediately by people with these diseases, the impact extends beyond them to include the healthcare system and wider society in terms of, for example, the costs and legal framework that are required to ensure good governance around organ donation These impacts vary between countries, depending, inter alia, on access to resources and societal attitudes It is important to investigate and compare impacts across contexts in order to improve our understanding This PhD examined distinct aspects of these impacts across contexts using a variety of analytic approaches

Chapter 2 examined the relationship between health-related quality of life (HRQoL) and CKD severity, and the consequent economic impact due to reduced HRQoL among a representative sample of community dwelling adults in the United Kingdom The evident decline in HRQoL among people even with milder stages of CKD underscored the importance of primary prevention, early diagnosis and secondary prevention in reducing the impact of CKD

As ESRD is expensive and access to renal replacement therapies (RRTs) is limited in many circumstances, there is considerable potential for disparities to occur in relation

to the management of ESRD Chapter 3 examined disparities among ESRD inpatients taking into account changes in the policy context in the United States of America Significant racial disparities regarding recorded anaemia, a modifiable and common complication of CKD, were observed Not only did Native Americans as a group experience the worst outcomes but their relative position declined after changes to policy Chapter 3 underlined unintended consequences of policy changes that may have been preventable

Complementing global efforts to address organ shortage, Chapter 4 examined attitudes (conceptualised in terms of passive support) and behaviours (conceptualised in terms

of active support) related to tissue donation after death among citizens from 27 European countries Here, a mismatch and unobserved heterogeneity between passive

iii

and active support for tissue donation, was observed which underscored the importance of examining both types of support together in order to attain a more complete understanding about attitudes and how, potentially, they shape behaviours Chapter 4 found the potential of incorporating both selfless and selfish motivations to translate latent passive into active support for donation

Finally, despite the evident medical benefits of living donor kidney transplantation (LDKT), there is limited information about the economic impact of LDKT in the UK The final chapter reviewed the evolution of the LDKT program in Northern Ireland, which currently leads the world in terms of living donor rate, and considered the implications of the program’s evolution for future cost-effectiveness studies Using the linked data provided by the Belfast Health and Social Care (HSC) Trust and the Northern Ireland HSC Business Service Organization, Chapter 5 demonstrated the superiority of LDKT over deceased donor kidney transplantations in terms of survival outcomes Furthermore, it demonstrated how this superiority persisted, despite the increasing complexity of case mix over time The chapter concluded by presenting a road map for a subsequent full economic evaluation of the LDKT program

In summary, the PhD thesis identified a number of economic aspects of CKD and ESRD including economic impact, equity in care, potential approaches to address the organ shortage crisis and the implication of program evolution for cost-effectiveness analyses

iv

ACKNOWLEDGEMENT

First and foremost, I would like to express my greatest gratitude to my two supervisors Prof Ciaran O’Neill and Prof Michael Donnelly You are the best supervisors I could ever wish for, thank you for your invaluable guidance and immeasurable support, your encouragement and patience, and your positive energy that have fulfilled my PhD journey What you have taught me sets a steady foundation for my future career

I am also grateful to the late Prof Liam Murray who opened the door for me to Queen’s and gave me the great support since my first step in this journey To Ms Heather Taylor and the CPH Admin team, thank you for your continuous support

My thesis as well as the other research I have been involved in have been considerably improved thanks to many of the collaborators I would like to send especial thanks to Prof Peter Maxwell, Dr Aisling Courtney, Dr Michael Quinn, and Dr Grainne Crealey for your expertise and advice

I also wish to thank everyone in Centre for Public Health, the Cancer Epidemiology and Health Services Research Group, and the Honest Broker Service team who provided me with support and advice throughout my study

To my friends at Centre for Public Health, NUI Galway, CEM, and the Vietnamese community at Belfast, thank you for making my everyday enjoyable and making Belfast

my second home Special thanks to Jinnan, Haydee, KimTu, Ngan, Anna, Luke, Dan, Danielle, Ethna, Leonie, and Euan To my friends and colleagues in Vietnam, thank you for keeping me posted with continuous changes in my home country

To my parents, my parents-in-law, and my three handsome brothers, thank you for your unconditional and infinite love which have shaped me the way I am today To my Mom, thank you for inspiring me to become a strong and independent woman

I am also grateful for my funding VIED and my University in Vietnam for giving me the opportunity to pursue my PhD at Queen’s

To the examiners, thank you for giving me such a friendly and helpful viva, for your comments and advice that improved the quality of my thesis

Finally, to my husband, your incredible support, inspiration, patience, and your love have lifted me up and sparked joy to my life I am so lucky to have you accompany

me in this challenging but colourful journey Thank you for everything!

v

TABLE OF CONTENTS

DECLARATION i

ABSTRACT ii

ACKNOWLEDGEMENT iv

TABLE OF CONTENTS v

LIST OF TABLES xi

LIST OF FIGURES xii

LIST OF APPENDICES xv

DISSEMINATION OF RESULTS xvi

LIST OF ABBREVIATION xvii

Chapter 1 INTRODUCTION 1

1.1 CHRONIC KIDNEY DISEASE 2

1.1.1 Kidney function 2

1.1.2 Chronic kidney disease definition and classification 5

1.1.3 Epidemiology 7

1.1.4 CKD management 15

1.2 END-STAGE RENAL DISEASE 20

1.2.1 End-stage renal disease definition 20

1.2.2 Epidemiology 21

1.2.3 ESRD management 28

1.3 ECONOMIC IMPACT OF CKD AND ESRD 38

1.3.1 Mortality 39

1.3.2 Morbidity 41

1.3.3 Healthcare cost of CKD and ESRD 43

1.4 AIM AND OBJECTIVES OF THIS THESIS 46

1.4.1 Aim of this thesis 46

vi

1.4.2 Objectives of this thesis 46

1.4.3 Structure of this thesis 48

Chapter 2 HEALTH-RELATED QUALITY OF LIFE AMONG PATIENTS WITH CHRONIC KIDNEY DISEASE AND THE ASSOCIATED ECONOMIC IMPACT IN THE UNITED KINGDOM 49

2.1 INTRODUCTION 49

2.1.1 Chronic kidney disease 49

2.1.2 Health-related quality of life 50

2.1.3 Aim of the study 52

2.2 METHODS 52

2.2.1 Data source 52

2.2.2 Statistical analysis 53

2.2.3 Sensitivity analysis 55

2.2.4 Projecting economic burden of CKD 56

2.3 RESULTS 57

2.3.1 Descriptive statistic 57

2.3.2 Base case analysis 60

2.3.3 Sensitivity analysis 61

2.3.4 The relationships between CKD status and the specific EQ-5D domains of health 65

2.3.5 The burden of chronic kidney disease and projections among those with diabetes to 2025 66

2.4 DISCUSSION 68

2.4.1 Strengths 70

2.4.2 Limitations 71

2.4.3 Implication for future study 72

2.5 CONCLUSION 73

vii

Chapter 3 INPATIENT CARE FOR END-STAGE RENAL DISEASE

PATIENTS IN THE UNITED STATES 74

3.1 INTRODUCTION 74

3.1.1 Anaemia and policy changes related to anaemia treatment 76

3.1.2 Depression 78

3.1.3 Mortality 79

3.1.4 Hospital incurred costs 80

3.1.5 Aim of the study 81

3.2 METHODOLOGY 82

3.2.1 Data source and study population 82

3.2.2 Study outcomes 83

3.2.3 Study variables 84

3.2.4 Statistical analyses 85

3.2.5 Sensitivity analyses 91

3.3 RESULTS 92

3.3.1 Descriptive statistics 92

3.3.2 Racial/ethnic disparities in anaemia complication 98

3.3.3 Depression 102

3.3.4 Inpatient mortality 104

3.1.2 Discharge destination 107

3.3.5 Length of stay in the hospital 109

3.3.6 Hospital incurred costs 110

3.3.7 Sensitivity analyses 112

3.4 DISCUSSION 113

3.4.1 Changes in the number of hospitalizations with ESRD over time 113

3.4.2 Racial/ethnic disparities 114

3.4.3 Depression among inpatients with ESRD 119

viii

3.4.4 Inpatient mortality 121

3.4.5 Costs and length of stay in the hospital 122

3.4.6 Strengths 124

3.4.7 Limitations 124

3.5 CONCLUSION 126

Chapter 4 UNDERSTANDING SUPPORT FOR TISSUE DONATION ACROSS 27 EUROPEAN COUNTRIES 128

4.1 INTRODUCTION 128

4.1.1 Organ and tissue donation 128

4.1.2 Donation legislation 130

4.1.3 Motivations and reservations 131

4.1.4 Aim of the study 133

4.2 METHODS 134

4.2.1 Data source 134

4.2.2 Study variables 134

4.2.3 Statistical analyses 137

4.3 RESULTS 140

4.3.1 Characteristics of the study cohort 140

4.3.2 The unobserved heterogeneity in the correlation between passive and active support for tissue donation 144

4.3.3 The association of legal context and support for tissue donation 145

4.3.4 The association of motivational contexts and support for tissue donation 147

4.3.5 Age generation and support for tissue donation 149

4.4 DISCUSSION 150

4.4.1 Limitations 156

4.5 CONCLUSIONS 157

ix

TRANSPLANTATION IN NORTHERN IRELAND: IMPLICATIONS FOR

COST-EFFECTIVENESS STUDIES 159

5.1 INTRODUCTION TO KIDNEY TRANSPLANTATION 160

5.1.1 Renal replacement therapy 160

5.1.2 Living donor kidney transplantation 162

5.1.3 Transplant activity in the UK 171

5.2 THE EVOLUTION OF THE LIVING DONOR TRANSPLANTATION PROGRAM IN NORTHERN IRELAND AND IMPLICATIONS FOR COST-EFFECTIVENESS ANALYSES 174

5.2.1 Kidney transplant activity in Northern Ireland 174

5.2.2 The evolution of living donor kidney transplant program in Northern Ireland since 2010 176

5.2.3 A study of the transplant population in Northern Ireland from 2010 to 2017 177

5.2.4 Implications for cost-effectiveness analyses 187

5.3 EQUALITY IN LIVING DONOR KIDNEY TRANSPLANTATION 192

5.3.1 Socio-economic disparities 193

5.3.2 Ethical consideration 198

5.4 POTENTIAL FUTURE RESEARCH 199

5.5 CONCLUSION 202

Chapter 6 DISCUSSION AND CONCLUSION 203

6.1 SUMMARY OF MAIN FINDINGS 203

6.1.1 The impact of early stages CKD 203

6.1.2 The potential unintended consequences of policy 205

6.1.3 Kidney transplantation programs 207

6.2 POLICY IMPLICATIONS 210

6.2.1 CKD and ESRD prevention 210

x

6.2.2 Policy related to CKD and ESRD management 211

6.2.3 Policy aimed at promoting kidney transplantation 212

6.3 LIMITATIONS 214

6.4 FUTURE RESEARCH 216

6.5 OVERALL CONCLUSIONS 219

REFERENCE 221

APPENDIX 288

APPENDIX 1 Supplementary data for Chapter 2 288

APPENDIX 2 Study population and detailed results from regression models obtained from HCUP-NIS data 290

APPENDIX 3 Variable definition based on the Eurobarometer survey 321

xi

LIST OF TABLES

Table 2.1 Demographic characteristics, socioeconomic status and comorbidities of

HSEs 2010 participants, stratified to eGFR level 58

Table 2.2 Sensitivity analysis results 62

Table 2.3 The estimated average marginal effects of CKD severity on different EQ-5D domains of health 65

Table 3.1 Characteristics of the pooled cohort from 2008 to 2016 95

Table 3.2 The relative likelihood of inpatient CKD-related anaemia among ESRD patients by races/ethnicities 99

Table 3.3 The association of PPS/ACA and racial/ethnic disparities 102

Table 3.4 Likelihood of depression among ESRD patients undergoing different RRTs 103

Table 3.5 Inpatient mortality in the pooled cohort 104

Table 3.6 The likelihood of discharge to a healthcare facility 108

Table 3.7 Length of stay in hospital 110

Table 3.8 Hospital incurred costs 111

Table 4.1 Sample characteristics 141

Table 4.2 The correlation coefficients among the residuals of different biprobit models 145

Table 5.1 Summary of studies on cost of living donor kidney transplantation (all costs are presented in 2019 US dollar) 164

Table 5.2 Cox Proportional hazard regression results 186

xii

LIST OF FIGURES

Figure 1.1 Prognosis of CKD by GFR and albuminuria category (KDIGO 2012 guideline) [6] 7Figure 1.2 GBD (Global burden disease study) estimated incidence of CKD cases by regions from 1990 to 2017 (Adapted from the GBD study 2017 [32]) 9Figure 1.3 GBD estimated prevalence of CKD cases by regions from 1990 to 2017 (Adapted from the GBD study 2017 [32]) 10Figure 1.4 The prevalence of CKD by stage among NHANES participants, 2001-2016 [30] 11Figure 1.5 The distribution of CKD stages based on eGFR and urine albumin, by age and sex (HSEs 2016) [38] 12Figure 1.6 Conceptual model of public health approach for CKD management [62] 15Figure 1.7 Annual incidence and prevalence rates of ESRD in different countries [23] 22Figure 1.8 The incidence rates of ESRD, by selected country, 2000–2012 [134] 23Figure 1.9 Trends in the (a) crude and standardized incidence rates of ESRD, and (b) the annual percentage change in the standardized incidence rate of ESRD in the U.S population, 1980-2016 [30] 24Figure 1.10 Trends in the prevalence of ESRD in the US population, 1980-2016 [30] 25Figure 1.11 RRTs incidence rates in the countries of the UK from 1990 to 2016 [138] 25Figure 1.12 RRT prevalence rates by age group and UK country until 31/12/2016 (pmp: per million population) [138] 26Figure 1.13 Adjusted incidence and prevalence per million population by country/region in 2016 (Adapted from ERA-EDTA Registry annual report 2016 [139]) 27

xiii

Figure 1.14 Trends in the annual number of ESRD (a) incident cases and (b) prevalent cases, by modality, in the US population, 1980-2016 [30] 29Figure 1.15 Estimated number of premature deaths due to the kidney shortage in the

US [195] 33Figure 1.16 Kidney transplants per million population by donor type and by country/ region, unadjusted (Adapted from ERA-EDTA Registry Annual Report 2016 [139]) 34Figure 1.17 Temporal trends in the annual number of living kidney donation transplants [208] 38Figure 1.18 GBD estimated death rates attributed to CKD from 1990 to 2017 [32] 40Figure 1.19 Adjusted all-cause mortality by ESRD treatment modality from 2001 to

2016 in the US [30] 41Figure 2.1 The proportion of respondents reporting any problems in each domain of EQ-5D questionnaire 60Figure 2.2 Multivariable Tobit model exploring the relationship between CKD severity and HRQoL (the base case model) 61Figure 2.3 Factors associated with increased/decreased HRQoL (Model 2) 64Figure 2.4 Factors associated with increased/decreased HRQoL (Model 3) 64Figure 2.5 The projected economic burden of stages 3-5 CKD from 2012 to 2025 in the UK in patients with diabetes 67Figure 2.6 The projection of economic burden of CKD stage 3, stage 4 and stage 5 from 2012 to 2025 in the UK in patients with diabetes, by gender 67Figure 3.1 Temporal trends in the number of admissions with ESRD from 2008 to

2016, by RRTs 93Figure 3.2 Temporal trends (from 2008 to 2016) in the percentage ESRD admissions with depression, anaemia, or who died in hospital (by RRTs) 93Figure 3.3 Factors associated with the likelihood of receiving a kidney transplant procedure, HD, PD or no RRTs 97

xiv

Figure 3.4 Percentage of patients with recorded anaemia from 2008 to 2016, by race/ethnicity 98Figure 3.5 The predicted probability of anaemia from the fully adjusted logistic regression model across races/ethnicities compared to Whites 100Figure 3.6 The predicted probability of anaemia from the fully adjusted logistic regression model across gender, insurance types, income levels, and RRT types over time 101Figure 3.7 The association of RRTs and inpatient mortality across age groups and over time 106Figure 3.8 Inpatient mortality among Native and Black Americans compared to White Americans across age groups 107Figure 3.9 The association between receipt of RRTs and discharge destination across age groups 109Figure 4.1.Variable definition process 136Figure 4.2 Levels of support for donation across countries 143Figure 4.3 The association of cohort characteristics and support for tissue donation (passive and active support) 146Figure 4.4 The association of respondents’ characteristics and passive support for tissue donation in three motivational subgroups 147Figure 4.5 The association of respondents’ characteristics and active support for tissue donation in three motivational subgroups 148Figure 4.6 Passive and active support for tissue donation of different generations compared to “silent” generation, stratified by motivations 149Figure 5.1 Adult living donor kidney transplants in the UK from 1 April 2004 to 31 March 2019 [606] 171Figure 5.2 The number of LDKT and dialysis patients in Northern Ireland (2000-2015), adopted from Graham et al (2018) [196] 175Figure 5.3 Age distribution of recipients and donors (stratified by time period) 180

xv

Figure 5.4 RRT modality and primary renal diagnosis before transplantation (stratified by LDKT/DDKT and time period) 181Figure 5.5 Number of HLA-ABDR mismatched (stratified by LDKT/DDKT and time period) 182Figure 5.6 Time on RRT prior to transplant and LOS in hospital for the transplant surgery (stratified by LDKT/DDKT and time period) 183Figure 5.7 Kaplan-Meier survival curves for LDKT and DDKT 185Figure 5.8 Kaplan-Meier survival curves for cohort 2010-2013 and 2014-2017 185

LIST OF APPENDICES

APPENDIX 1 Supplementary data for Chapter 2 288APPENDIX 2 Study population and detailed results from regression models obtained from HCUP-NIS data 290APPENDIX 3 Variable definition based on the Eurobarometer survey 321

PLoS One 2018;13(11):e0207960

Nguyen NTQ, Maxwell AP, Donnelly M, O’Neill C The Role Of Motivational And Legal Contexts In Understanding Support For Tissue Donation Across 27 European

Countries European Journal of Public Health 2020;(ckaa148):1–6

Nguyen NTQ, Maxwell AP, Donnelly M, Neill CO Prospective payment system and racial / ethnic disparities : a national retrospective observational study in anaemia

complication among end-stage renal disease patients in the US BMC Nephrology

2020;21(423):1–9

Conference presentations

Nguyen NTQ, Cockwell P, Maxwell AP, Griffin M, Brien TO, Neill CO PUK17 − The Relationship Between Severity of Chronic Kidney Disease and Health-Related Quality of Life Among a Nationally Representative Sample of Community Dwelling

Adults in England Value in Health 2017;20(20):A491

Nguyen NTQ, Maxwell P, Donnelly M, O’Neill C PUK21 Racial Disparities In Recording Anemia As A Complication Among End-Stage Renal Disease Patients

Value in Health 2019;22:S384–5

xvii

LIST OF ABBREVIATION

ACCI Age-adjusted Deyo-Charlson Comorbidity Index

ACR Albumin/creatinine ratio

AHRQ Agency for Healthcare Research and Quality

AIC Akaike information criterion

ANOVA Analysis of variance

BIC Bayesian information criterion

CCI Deyo-Charlson Comorbidity Index

CKD-EPI Chronic Kidney Disease Epidemiology Collaboration equation

DALYs Disability-adjusted life years

DDKT Deceased donor kidney transplantation

DiD Difference-in-difference

eGFR Estimated glomerular filtration rate

ERA-EDTA European Renal Association – European Dialysis and Transplant

Association registry ESAs Erythropoiesis stimulating agents

ESRD End-stage renal disease

FDA Food and Drug Administration

GBD Global burden disease study

GFR Glomerular filtration rate

GLMs Generalised linear models

HCUP-NIS National (Nationwide) Inpatient Sample - Healthcare Cost and

Utilisation Project

HRQoL Health-related quality of life

HSEs Health Surveys for England

ICD-10 International Classification of Diseases 10th Revision

ICD-9 International Classification of Diseases 9th Revision

xviii

IRODaT International Registry In Organ Donation And Transplantation KDIGO Kidney Disease: Improving Global Outcomes

LDKT Living donor kidney transplantation

LMIC Low- and middle-income countries

LOS Length of stay in the hospital

MDRD Modification of Diet in Renal Disease equation

NHANES American National Health and Nutrition Examination Survey

NICE National Institute for Health and Care Excellence

QALY Quality-adjusted life year

RRTs Renal replacement therapies

USRDS United States Renal Data System

YLDs Years lived with disability

YLLs Years of life lost

an estimated 697.5 million people – equivalent to 9% of global population – were diagnosed with CKD [1] CKD ranked as the 16th leading cause of global years of life lost (YLLs) and accounted for more than 1.2 million deaths in 2017 [2] However, it

is frequently unrecognised, co-existing with other health conditions Awareness about CKD is alarmingly low – almost 90% of patients were unaware of their kidney disease status [4] In most cases, CKD is a ‘silent killer’ because it tends to be asymptomatic until the late stages of the disease when there is significant loss of kidney function However, increasingly, there are international collaborative efforts aimed at early detection and delaying progression of CKD, for example, the National Kidney

Foundation published the Kidney Disease Outcomes Quality Initiative Guidelines in

2002 and the updated Kidney Disease: Improving Global Outcomes (KDIGO) in 2012

[5,6] Since then, CKD has gained increasing recognition in clinical practice and research However, this area of research remains relatively neglected in terms of studies and receives limited funding compared to other diseases [7]

Given the global health challenge caused by CKD epidemic, greater multidisciplinary cooperation between healthcare professionals, healthcare providers, researchers, and patients is required to help shape and pursue a research agenda in this area The World

Kidney Day 2020’s theme of ‘Kidney health for everyone everywhere – from

Prevention to Detection and Equitable Access to Care’ calling for early prevention

and universal, sustainable and equitable treatment for kidney disease provides one example of attempts to raise awareness of the issue [8] To raise awareness though relevant stakeholders should be informed with data on the impact of CKD on the society and economy; on the effectiveness and cost-effectiveness of treatments to manage the condition, as well as public awareness and attitudes toward disease management

2

The research contained in this PhD sought to improve understanding in this area by examining economic aspects of kidney disease and its management using several analytical approaches The first section of Chapter 1 provides an overview of the kidney and the measurement of kidney functioning, CKD definition and classification, epidemiological data regarding CKD, and CKD progression and complications management The second section of this chapter focuses on late stage CKD – end-stage renal disease (ESRD) – by providing an overview of the epidemiology and management of ESRD These two sections set out the context of the PhD research and are intended to help readers understand the importance of examining the economic impact of CKD and ESRD in particular The third section of the chapter presents data about the economic impact of CKD and ESRD in terms of mortality, morbidity and the healthcare costs that are associated with CKD and ESRD and their management

In addition, section 3 presents and discusses the role of donation/transplantation and the context in which decisions around donation are made Finally, section 4 concludes Chapter 1 by presenting the aims and objectives of the PhD thesis

1.1 CHRONIC KIDNEY DISEASE

1.1.1 Kidney function

The kidneys are a vital pair of bean-shaped organs that are responsible for the excretion

of waste products, control of body homeostasis, and endocrine and metabolic functions [9] Toxic substances are converted to soluble forms via metabolism in liver before being excreted by kidneys into the urine By selectively excreting or reabsorbing different compounds such as water, sodium, potassium, calcium, magnesium and others, kidneys ensure the chemical balance of the blood and maintain the body’s homeostasis Finally, the endocrine and metabolic functions of kidneys are vital Kidneys produce erythropoietin (an important hormone to promote the formation

of red blood cells by the bone marrow), produce renin (an enzyme that plays an important role in controlling blood pressure via the renin–angiotensin–aldosterone hormonal system), and activate vitamin D (to maintain strong bones and regulate the immune system) Kidneys also respond to other functional hormones such

as aldosterone, prostaglandins, cortisol, parathyroid hormone and calcitonin [9]

3

The basic functional and structural unit of the kidney is a nephron which comprises a vascular portion (glomerulus and renal venous plexus) and a tubular portion (renal tubule) [9] Each kidney has approximately one million nephrons When the kidney is impaired, the number of functioning glomeruli and the efficiency of these glomeruli both decrease, which leads to the abnormal filtration and reabsorption function of the kidneys As a result, some substances that are normally excreted now appear in the blood, and other substances that are normally unpresented in urine now appear in the urine The overall health of the kidneys can be evaluated by measuring unusual blood level or urine level of these substances Other tests including an ultrasound scan, magnetic resonance imaging scan, computerised tomography scan, and kidney biopsy are also used in clinical practice

1.1.1.1 Glomerular filtration rate

The glomerular filtration rate (GFR) has been long recognised as the best indicator of renal function and is critical for diagnosing kidney impairment, progression of kidney disease, and informing medication use among patients with malfunctioned kidneys [10] GFR is used to measure the amount of blood filtered in a given minute by kidneys and is approximately proportional to the number of nephrons and the size of the glomeruli The normal value of GFR is 90 mL/min/1.73m2 and above GFR decreases with age by a rate of 0.4 to 1.2 mL/min/1.73m2 per year [11]

As GFR cannot be measured directly, alternative methods have been proposed and used in clinical practice The gold standard for GFR measurement is renal inulin clearance, though this method is merely used in confirmatory diagnosis due to its cost, difficult assays and inconvenience in collecting timed urine samples When the renal clearance of a substance is equal to its plasma clearance, the single-injection technique can be used to measure GFR by monitoring the rate of substance disappearing from plasma [12] Several exogenous markers could be used in this case including some radioisotopes such as 96mTc-DTPA (diethylene triamine penta-acetic acid), 51Cr-EDTA (ethylene diamine tetraacetic acid), and 125I-iothalamate However, these methods are used only occasionally in clinical practice due to their costs, inconvenience and the exposure to radioactivity [9]

The most common method to estimate GFR is creatinine clearance-based estimates [9] Creatinine is a waste product produced by the breakdown of creatine in the muscle

4

at a relatively constant rate Normally, the kidneys filter creatinine from blood into urine and reabsorb a negligible portion of it Kidney function can be assessed by measuring creatine clearance - the amount of creatine-free blood filtered by kidney in

a given minute

1.1.1.2 Equations to estimate glomerular filtration rate

Two main methods to measure creatinine clearance are collecting 24-hour urine sample and estimating single blood level of creatinine One major downside of the former method is the need to accurately collect urine over 24 hours This practice requires toilet trained and the strict adherence of patients Therefore, several equations have been developed to predict the daily excretion rate of creatinine using its blood level and other factors such as body size, sex, age, and races/ethnicities, which could obviate the need for 24-hour urine collection [9]

1.1.1.2.1 Cockcroft and Gault equation

The Cockcroft and Gault equation was proposed in 1976 to estimate creatinine clearance and has been widely used in clinical practice and research [13]:

𝑒𝐶𝑟𝐶𝑙 = (140 − 𝑎𝑔𝑒)𝑥 (𝑤𝑡 𝑖𝑛 𝑘𝑔)𝑥 (0.85 𝑖𝑓 𝑓𝑒𝑚𝑎𝑙𝑒)

72 𝑥 𝑆𝐶𝑟 𝑖𝑛 𝑚𝑔/𝑑𝐿Or:

𝑒𝐶𝑟𝐶𝑙 = (140 − 𝑎𝑔𝑒)𝑥 (𝑤𝑡 𝑖𝑛 𝑘𝑔)𝑥 (0.85 𝑖𝑓 𝑓𝑒𝑚𝑎𝑙𝑒)

0.814 𝑥 𝑆𝐶𝑟 𝑖𝑛 𝑚𝑐𝑚𝑜𝑙/𝐿Wherein: eCrCl denotes the estimated clearance of creatinine, SCr denotes the serum creatinine level, wt denotes weight in kilogram, mg/dL denotes milligram per decilitre, mcmol/L denotes micromoles per litre

1.1.1.2.2 Modification of Diet in Renal Disease Equation (MDRD)

The MDRD equation was developed during Modification of Diet in Renal Disease (MDRD) Study in a sample of patients mostly having GFR < 60 mL/min/1.73m2 [14] GFR is measured using isotopically tagged iothalamate While GFR > 60 mL/min/1.73m2, the MDRD equation provides less reliable results

The isotope dilution mass spectrometry traceable MDRD equation:

𝑒𝐺𝐹𝑅/1.73 𝑚2 = 175𝑥 𝑆𝐶𝑟−1.154 𝑥𝑎𝑔𝑒−0.203𝑥[1.21 𝑖𝑓 𝑏𝑙𝑎𝑐𝑘]𝑥[0.742 𝑖𝑓 𝑓𝑒𝑚𝑎𝑙𝑒]

5

Wherein: eGFR is the estimated glomerular filtration rate, SCr is the serum creatine level

1.1.1.2.3 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation

The most recent equation is the CKD-EPI equation that based on a more representative sample compared to the previous equations This is a set of eight equations that could

be chosen based on gender (female/male), race (Caucasian/African American), and if serum creatinine (SCr) is in a lower or higher range CKD-EPI has proved to be more accurate and provide less bias results compared to MDRD equation, especially among those with estimated GFR (eGFR) over 50 mL/min/1.73m2 [15] The CKD-EPI equation also offered the ability to more accurately predict the risk for mortality and ESRD progression than MDRD equation across a broad range of populations [16,17] These equations have been used widely in practice However, in some specific populations such as those with acute kidney injuries, very lean or obese individuals, there should be caution when using equations to estimate GFR and it may be best to apply the traditional 24-hour urine method [9] In addition, though being the currently optimal equation, the use of CKD-EPI could result in higher estimates of GFR among young people, and lower estimates among the elderly, that could affect the CKD management from a population perspective given the different age structure in different community and the worldwide ageing population tendency [18]

1.1.2 Chronic kidney disease definition and classification

The National Kidney Foundation and Kidney Disease: Improving Global Outcomes (KDIGO) defined CKD as structural or functional abnormalities of the kidney lasting for more than 3 months, manifested by either kidney damage or GFR less than 60 mL/min/1.73m2 (with or without kidney damage) [19,20] The updated KDIGO

guideline in 2012 defined CKD as ‘abnormalities of kidney structure or function,

present for >3 months, with implications for health’ [6] The abnormalities comprise

decreased GFR level (< 60 mL/min/1.73 m2) and markers of kidney damage (including albuminuria, urine sediment abnormalities, electrolyte and other abnormalities due to tubular disorders, abnormalities detected by histology, structural abnormalities detected by imaging and history of kidney transplantation) [6]

6

Another severe clinical syndrome related to kidney impairment is acute kidney injury (AKI) A growing body of literature has reported the close relationship between CKD and AKI as they are likely to accelerate each other [21], although their potential causal relationship remains unclear [22] Distinct from CKD, the KDIGO 2012 defined AKI

as ‘an absolute increase in SCr, at least 0.3 mg/dL (26.5 μmol/L) within 48 hours or

by a 50% increase in SCr from baseline within 7 days, or a urine volume of less than 0.5 mL/kg/h for at least 6 hours’ Critically ill patients with AKI could also require

renal replacement therapies (RRTs) and are likely to develop worse long-term outcomes such as ESRD [21]

The 2012 KDIGO guideline suggested the classification of CKD based on cause, GFR category, and albuminuria category (CGA classification) GFR could be categorised

to five levels from G1 to G5, albuminuria could be categorised to three levels from A1

to A3 Common causes of CKD include diabetic kidney disease, idiopathic focal sclerosis, polycystic kidney disease, vesicoureteral reflex, distal renal tubular acidosis, hypertensive kidney disease, CKD presumably due to diabetes and hypertension, kidney transplant recipient, and CKD with unknown causes [6] CKD stage 1 is denoted by patients with unimpaired GFR (>90 mL/min/1.73m2) and persistent proteinuria or structural abnormalities CKD stages 2 to 5 were defined by the extent

to which GFR declines Figure 1.1 depicts the prognosis of CKD by GFR and albuminuria level from KDIGO 2012 guideline [6] All patients with GFR under 29 mL/min/1.73m2 (G4 or G5) are at high risk regardless of albuminuria level, while the prognosis of those with GFR level G3 depends on albuminuria level

7

Figure 1.1 Prognosis of CKD by GFR and albuminuria category (KDIGO 2012 guideline) [6]

1.1.3 Epidemiology

1.1.3.1 The global context

The combination of an aging population and an increasing epidemic of, respectively, diabetes mellitus, hypertension, and obesity have been recognised as key drivers of the significant rise in the global prevalence of CKD [23] Worldwide, in 2017, it was estimated that 697.5 million (649.2 to 752.0) people had diagnosed CKD among whom 19.7 million (17.7 to 22.0) had newly diagnosed CKD [1] Among prevalent cases, CKD was most common among people aged between 60-64 years, and the identified causes of CKD were mainly type 2 diabetes mellitus (18%), glomerulonephritis (4.1%) and hypertension (3.4%) [2] The estimated number of years lived with disability (YLDs) due to CKD was 7.3 million [1], which was mainly attributed to population growth and aging [24] CKD was ranked as the 16th leading cause of global years of life lost (YLLs) in 2017, a significant increase of 21% since 2007 [2] Regarding

1.1.3.1.2 Hypertension

Hypertension is both a common cause of CKD and a consequence of CKD [18,26] In

2017, it was estimated that 347.4 thousands of CKD deaths and 5.9 million YLLs were attributable to hypertension, which represents a 41.4% and 33.4% increase since 2007, respectively [2] Although it is present in 80% of CKD cases [26], awareness of hypertension and blood pressure control among CKD patients remains suboptimal [29] Fortunately, recent data from United States Renal Data System (USRDS) showed

a 20% decrease in the numbers of unaware hypertension and a 27% increase in the percentage of treated hypertension among CKD patients [30] The level of increased blood pressure is highly associated with CKD progression and the risk of cardiovascular disease (CVD) which has been elevated due to the progressive CKD [26] Therefore, treatment for hypertension could slow the progression of CKD and prevent CVD outcomes irrespective of causes of the disease As the central mechanism

of hypertension-related CKD is related to the activity of renin–angiotensin–aldosterone system, it is recommended to use angiotensin-converting-enzyme

9

inhibitors or angiotensin receptor blockers as first-line agents to control blood pressure and reduce proteinuria [31]

1.1.3.2 Prevalence and incidence of CKD

1.1.3.2.1 Prevalence and incidence of CKD by regions

There has been an evident upward trend regarding both prevalence and incidence of CKD across regions, with significantly steeper increase in East Asia and Pacific, and South Asia (Figure 1.2 and Figure 1.3) In 2017, East Asia and Pacific led the world

in both the incidence (4.4 million) and prevalence (235.6 million) of CKD This is followed by the emerging South Asia region with 40.9 million incident cases and 147.5 million prevalent cases, which took over the second ranking of Europe and Central Asia region in 2016 (incidence) and in 1999 (prevalence) North America constituted 1.6 million incident cases and 42.3 million prevalent cases in 2016 [32]

Figure 1.2 GBD (Global burden disease study) estimated incidence of CKD cases by regions from 1990 to 2017 (Adapted from the GBD study 2017 [32])

East Asia & Pacific Europe & Central Asia Latin America & Caribbean

Middle East & North Africa North America South Asia

Sub-Saharan Africa

10

Figure 1.3 GBD estimated prevalence of CKD cases by regions from 1990 to 2017 (Adapted from the GBD study 2017 [32])

1.1.3.2.2 Prevalence and incidence of CKD by stages

In a systematic review in 2016, the global prevalence of stages 1 to 5 CKD and stages

3 to 5 CKD were estimated to be 13.4% (95% confidence interval CI: 11.7–15.1%) and 10.6% (95%CI: 9.2–12.2%), respectively [33] A majority of CKD prevalence was stage 3 CKD which accounted for 7.6% (95%CI: 6.4–8.9%), followed by stage 2 CKD (3.9%), stage 1 CKD (3.5%), stage 4 CKD (0.4%), and stage 5 CKD (0.1%) [33] However, given the potential issues related to the comparability and accuracy of measurements across countries, especially low- and middle- income countries, care is warranted with respect to comparisons such as this This section focused on reporting CKD prevalence in the US, United Kingdom (UK) and European countries for a demonstrative purpose of the subsequent chapters

a The United States of America (US)

In a meta-analysis, Hill et al estimated the prevalence of stages 1 to 5 CKD in the US and Canada to be 15.5% (95%CI: 11.7-19.2%) and the prevalence of stages 3 to 5 CKD to be 14.4% (95%CI: 8.5-20.4%) [33] Data from the American National Health and Nutrition Examination Survey (NHANES) demonstrated that prevalence of stages

East Asia & Pacific Europe & Central Asia Latin America & Caribbean

Middle East & North Africa North America South Asia

Sub-Saharan Africa

11

3 to 4 CKD increased between the late 1990s to the early 2000’s but stabilized since 2003-2004 [34] In the former period, while higher prevalence of diabetes, hypertension and obesity contributed to the observed increase among stage 2 CKD, it was the shift to an older population that primarily explained the prevalent stage 3 CKD [35] Between 2003-2004 and 2011-2012, there was little difference when comparing CKD prevalence even after controlling for age, sex, ethnicity and diabetes mellitus status [34] Similarly, the USRDS 2018 also indicated that CKD prevalence has remained relatively stable during the last 20 years The overall prevalence of stages 1

to 5 CKD was reported to be 14.8% in 2013-2016, while stage 3 CKD was the most prevalent stage which accounted for 6.4% (Figure 1.4) [30] The prevalence of CKD among diabetic and hypertensive population also decreased over time and were reported to be 36% and 31% of these populations during 2013-2016 period, respectively [30] While providing useful estimates, however, NHANES surveys have

a relatively small sample size (ranging from 4009 in NHANES 1990-2000 to 15484 in NHANES 1988-1994) [34], thus making it difficult to report detailed data among specific minority group such as Native Americans (who accounted for roughly 2% of the US population)

Figure 1.4 The prevalence of CKD by stage among NHANES participants, 2001-2016 [30]

12

b United Kingdom

The population prevalence of CKD in England was reported for the first time in 2010 using Health Surveys for England (HSEs) 2009/2010 data [36] The overall study-defined prevalence of stages 3 to 5 CKD were 6% in males and 7% in females For stages 1 to 5 CKD, the study-defined prevalence was estimated to be 11-12% in males and 14-15% in females [36] On the other hand, the prevalence of stages 3 to 5 CKD based on the Quality Improvement in CKD trial was estimated to be 6.76% in general population (aged over 18), 9.11% in females, and 4.39% in males [37]

The most recent report on CKD using HSEs data was released in 2016 Among all adults aged 35 and over, the prevalence of stages 1 to 5 CKD was 15%, among whom 51.7% had stages 3 to 5 CKD CKD was still more prevalent among women (17%) than men (12%) (Figure 1.5) [38]

Figure 1.5 The distribution of CKD stages based on eGFR and urine albumin, by age and sex (HSEs 2016) [38]

13

c European countries

In 2016, the prevalence of stages 1 to 5 CKD in Europe was reported to be 18.4% (95%CI: 11.6-25.2%), and the prevalence of stages 3 to 5 CKD was estimated to be 11.9% (95%CI: 9.9-13.8%) [33] However, the prevalence of CKD varied considerably across countries in Europe [39] One study estimated that the adjusted prevalence of stages 1 to 5 CKD was estimated to range from 3.31% (95%CI: 3.30-3.33%) in Norway to 17.3% (95%CI: 16.5-18.1%) in northeast Germany [39] Similarly, the adjusted prevalence of more advanced stages 3 to 5 CKD was estimated

to vary from 1.0% (95%CI: 0.7-1.3%) in central Italy to 5.9% (95%CI: 5.2-6.6%) in northeast Germany [39] This observed pattern remained unchanged even after stratifying the sample to diabetes, hypertension and obesity status These variations were explained by multiple factors such as human and environmental differences, public health policies, genetic factors, heterogeneity in laboratory methods, and heterogeneity in study populations among the studied countries [39]

1.1.3.3 Risk factors and social determinants of CKD

Levey et al (2005) suggested four main groups of risk factors for CKD, including susceptibility factors, initiation factors, progression factors, and end-stage factors [20]

Susceptibility factors include those that could elevate the susceptibility of individuals

to kidney damage, such as older age, family history of CKD, minority racial/ethnic groups and low socio-economic status (SES) [20,26] The global estimates of CKD prevalence increased with age, ranging from 13.7% in 30s group to 34.3% in 70s group [33] Regarding more advanced stages 3 to 5 CKD, this pattern remained relatively stable, ranging from 8.9% in 30s group to 27.9% in 70s group [33] Family history could play a role as some kidney diseases are inherited such as polycystic kidney disease and Alport’ nephropathy [26] In addition, studies have found that low SES was associated with higher prevalence of CKD [40], greater risk of presenting with an advanced stage of CKD [41,42], increased risk of poor outcome and rapid CKD progression [42] Low-income groups also face barriers to healthy diets and lifestyle [43–45], which have been associated with favourable CKD outcomes [46] In low-income countries, under-nutrition and starvation due to food insecurity among low-income individuals could lead to children being born with low birth weight and related sequelae including CKD [47] In high-income countries, over-nutrition and bad dietary due to food availability among low-income ones again could lead to worse health

in Canada and Australia, and the South East-Asians in the UK Apart from the link to SES status, the increased risk of CKD among African population could be associated with their Apolipoprotein L1 risk variants [54,55]

Initiation factors refer to those that could directly initiate CKD such as diabetes,

hypertension, obesity, urinary tract infections, autoimmune disease, and recovery from acute kidney injury [20,26] Among those, diabetes is the leading cause of advanced CKD worldwide, followed by hypertension [6] Obesity is another important risk factor of CKD which might be independent of diabetes and hypertension [56,57] Evidence from systematic reviews showed that reducing weight among CKD patients could result in a reduction in proteinuria and blood pressure, and delay CKD progression [58] While diabetes and hypertension are the leading causes of CKD in most countries, glomerulonephritis and unknown causes are more prevalent in Asian and sub-Saharan African countries [23]

Progression factors are those that could worsen kidney damage, accelerate GFR

decline, or increase the risk of complications due to decreased kidney function [20,26] These include a higher level of proteinuria, higher blood pressure level, smoking, factors related to hypertension, anaemia, bone and mineral disorders and the use of nephrotoxic drugs such as cyclosporine or tacrolimus, lithium and non-steroidal anti-inflammatory drugs

End-stage factors refer to those that could increase morbidity and mortality in kidney

failures such as anaemia, low serum albumin, high serum phosphorus, CVD risk factors [20,26] Access to RRTs is also an important deciding factor for treatment outcomes among CKD patients [26]

Finally, some risk factors appear to be unique in some parts of the world These include exposure to herbal preparations and its nephrotoxicity in Asia and Africa; aristolochic-acid in Taiwan, China and Balkan endemic nephropathies; HIV-associated nephropathy

in Africa; and contaminated water in South Asia [23] Sex-related disparities in respect

15

of CKD outcomes have been reported While CKD is more prevalent in women than men [33] and women with diabetes have higher risk of incident CKD [59], men with CKD had higher risk of progression to ESRD and deaths [60,61]

1.1.4 CKD management

The increasing prevalence of diagnosed CKD and the number of patients requiring treatments of kidney failure has resulted in a huge amount of money being invested in the identification and management of the disease CKD has become a public health concern and management of CKD should be initiated even in early stages of the disease Crucially, CKD is preventable and ESRD could be delayed if appropriate preventative interventions are introduced to the right person at the right time The

World Kidney day theme in 2020 of ‘Kidney Health for Everyone Everywhere – from

Prevention to Detection and Equitable Access to Care’ emphasized the importance of

prevention at all levels, including primary prevention, secondary prevention, and tertiary prevention [8] The Centers for Disease Control and Prevention presented a conceptual model of the public health approach to CKD management (Figure 1.6) that illustrates three levels of prevention [62]

Figure 1.6 Conceptual model of public health approach for CKD management [62] Early identification of CKD seems to be the most cost-effective approach to manage the disease, delay progression and decrease risk of CVD [9] However, earlier stages

of CKD are often asymptomatic and mostly diagnosed during the evaluation of other conditions [6] The need for a population based screening program has been debated

in many countries wherein most evidence supported a targeted, opportunistic screening

16

program aimed at high risk population [63,64] For example, in the UK, the National Institute for Health and Care Excellence (NICE) recommended annual GFR monitoring for people who were prescribed nephrotoxic drugs (such as cyclosporine

or tacrolimus, lithium and non-steroidal anti-inflammatory drugs) [65] A test for CKD

is also offered for people with risk factors including diabetes, hypertension, acute kidney injury, cardiovascular disease, multisystem diseases with potential kidney involvement, structural renal tract disease, and family history of ESRD or hereditary kidney disease Nevertheless, health information and CKD guidance are not available

in all countries In a study of 125 countries on CKD management, only 9 countries (8%) had a renal registry for non-dialysis CKD and 62 countries (52%) had a CKD guideline [66]

After being initially diagnosed with CKD, patients should be tested repeatedly to exclude the possibility of acute kidney injuries and other treatable conditions such as nephrotic syndrome or vasculitis [9] Once CKD has been diagnosed, as a rule of thumb, reducing CVD risk should be prioritised among CKD stages 1 to 3a patients, and slowing progression to ESRD is the predominant strategy for CKD stages 3b to 4 [18] However, the review and management of patients depend not only on CKD severity but also on individual circumstances In order to reduce the risk of progression and CVD, there should be a combination of lifestyle modification, lipid lowering, glycaemic control, and blood pressure control Detection and management of CKD complications, medications review, progression prevention, and consideration of nephrologist referrals or RRTs referrals are of importance [9]

1.1.4.1 Prevent progression

The most serious outcome of CKD is kidney failure While some people could face a rapid progression and have kidney failure within months, in most of the cases, CKD is

chronic and prolongs over decades [18] Progression of CKD is defined as ‘a sustained

decrease in GFR of 25% or more and a change in GFR category within 12 months’ or

‘a sustained decrease in GFR of 15 ml/min/1.73 m2 per year’ [6] In order to slow

down CKD progression, it is important to identify factors associated with CKD progression to inform prognosis Researchers have reported some factors such as cause

of CKD, level of GFR, level of albuminuria, age, sex, race/ethnicity, elevated blood pressure, hyperglycaemia, dyslipidaemia, smoking, obesity, history of cardiovascular disease, and ongoing exposure to nephrotoxic agents that might influence the

17

progression of the disease [6] The KDIGO 2012 guideline also informed 22 recommendations to delay CKD progression from lifestyle modification, dietary intake to medication uses and cardiovascular risk reduction in both diabetic and non-diabetic population [6]

1.1.4.2 Complications of chronic kidney disease

Regardless of region, age or aetiology, three main groups of complications of CKD are drug toxicity (due to altered pharmacokinetics of drugs as a result of malfunctioned kidney), metabolic and endocrine complications (comprising anemia, acidosis, bone and mineral disorders, and malnutrition), and increased risk of cardiovascular diseases and death [6]

1.1.4.2.1 Altered pharmacokinetic of drugs

Patients with CKD should be monitored for any signs of unexpected drug toxicity due

to their abnormal reaction to a number of drugs [67,68] Because kidney disease has multiple effects on pharmacokinetics of medications, failure to amend dosage following the severity of kidney disease could lead to adverse events of drugs and predispose patients to treatment failure [69] Although pharmacokinetics are well described in any drug information submitted to the US Food and Drug Administration (FDA), not all of them provide sufficient data on renal impaired populations [70] Some examples of medications that require cautious consideration are lithium (due to predominant excretion via kidney and severe neurotoxicity) [71], digoxin (due to narrow therapeutic index and accumulative feature) [72], cyclophosphamide (due to the active metabolite mainly eliminated by the kidney) [73], and metformin (contraindicated in patients with CKD due to concerns about lactic acidosis) [74,75] In the era of personalised medicine,

it is essential to consider not only drug clearance via kidney but also non-renal clearance, volume of distribution and bioavailability to adjust drug or its active metabolite concentration [76]

1.1.4.2.2 Anaemia

Since the early 19th century, anaemia has been reported as a common complication of CKD [77] Moderate and severe kidney dysfunction are strongly associated with a higher prevalence of anaemia and lower haemoglobin (Hb) level [78] Due to the reduced oxygen delivery to body’s organs and tissues, anaemia causes such symptoms

as fatigue, shortness of breath, insomnia, and headaches [79] Erythropoietin

18

deficiency due to impaired kidney function is one of the main causes of anaemia due

to CKD Other important causal factors are uremic-induced inhibitors of erythropoiesis, shortened erythrocyte survival, and disordered iron homeostasis [80] With the introduction of recombinant human erythropoietin and the related erythropoiesis stimulating agents (ESAs) in 1980s, anaemia management has obtained significant achievements for improving patients’ symptoms and freeing them from the complications of blood transfusions [81,82] However, studies have shown several pitfalls of ESAs such as not improving mortality or hospitalization outcomes or not delaying progression to ESRD [83], while concurrently increasing risk of death, adverse cardiovascular events, and stroke [84–86] Between different versions of ESAs, long-acting ESAs were also associated with a higher relative risk of mortality compared to short-acting ESAs [87] As a result, the previous Hb target range (from

10 to 12 g/dL) in anaemia treatment using ESAs was replaced by a cut-off of 11 g/dL

by FDA in 2011 and by KDIGO Anemia Work Group guideline in 2013 [88] In addition, approximately 10-20% of anaemia patients are resistant to ESAs treatments, thus other anaemia mechanism pathways have been investigated to develop optimal therapies [83] In fact, the disordered iron homeostasis has been increasingly recognised as a major contributor to the anaemia due to CKD [80] and many patients still required iron supplement to achieve adequate Hb level concurrently with ESAs [89] Given that anaemia is associated with many negative outcomes among CKD population [79], the impact of this condition is considerable

19

bone mineral content [92] As a result, metabolic acidosis is associated with progressive CKD, bone demineralization, skeletal muscle catabolism, and increased mortality Among transplanted patients, metabolic acidosis is also a frequent complication and could be an important risk factor for graft failure and mortality [94]

In order to alleviate the impact of metabolic acidosis on CKD patients, oral alkali is recommended by current clinical guidelines [95,96] A diet with reduced intake of acid-producing and increased intake of base-producing food and beverage is also recommended [95,96]

1.1.4.2.4 Bone and mineral disorders

Bone and mineral disorders are among the most difficult and time-consuming complications of CKD [6] Although the adaptive mechanism of the body in response

to kidney malfunction begins in stage 2, it is only at stage 3 CKD with more than 50% loss of kidney function that bone and mineral disorders become apparently symptomatic [97–99] This condition includes secondary hyperparathyroidism, hyperphosphatemia, decreased intestinal calcium absorption and disordered vitamin D metabolism [100] The compensatory mechanism of releasing calcium from bone as a result of metabolic acidosis could worsen bone and mineral disorders status [97] Management of hyperphosphatemia and prevention or treatment of secondary hyperparathyroidism remains the mainstay treatment of bone and mineral disorders in CKD patients [100]

1.1.4.2.5 Malnutrition

Elevated risk of malnutrition is common in patients with CKD, which comprises protein energy wasting and micronutrient deficiency [101,102] All pathophysiologic abnormalities in CKD patients could be associated with malnutrition Protein energy wasting affects up to 31% of adults with CKD and cannot be corrected by food intake solely [103] Factors such as anorexia, increased protein catabolism, decreased anabolism, systemic inflammation, metabolic acidosis and hormonal derangements could link to protein energy wasting [102,104–106] In addition, the decreased appetite and nutrient intake due to dietary restrictions, poor intestinal absorption, and metabolic acidosis predispose CKD patients at risk of micronutrient deficiencies [107] Similar

to other CKD complications, malnutrition has been found to increase the risk of morbidity, mortality and overall disease burden [101,102,108]

20

1.1.4.2.6 Cardiovascular disease and death

Cardiovascular disease (CVD) risk is pronounced in all stages of CKD [109–111] In the US, the prevalence of CVD is almost double among older patients with CKD (64.5%) compared to the general population (32.4%) [30] Research found that CVD

is most evident among patients with stage 3b to stage 4 CKD and those undergoing renal replacement therapies [112] The fact that CKD and CVD share common risk factors (such as diabetes mellitus, hypertension, smoking, family history, physical inactivity, dyslipidaemia) makes their relationship complicated and is referred to as cardiorenal syndrome [30,113] Apart from traditional risk factors of CVD, CKD patients are also exposed to other uraemia related risk factors such as anaemia and bone and mineral disorders [109] Reduced GFR remains a strong, independent factor

of CVD morbidity and mortality [114] In fact, CVD is the leading cause of death in patients with CKD [100] A meta-analysis reported that compared to normal eGFR level, the estimated odds ratio of death for eGFR levels of 80, 60 and

40 mL/min/1.73m2 were 1.9, 2.6 and 4.4, respectively [115] Patients with CKD are more likely to die due to CVD than to progress to ESRD [110,116], therefore managing this complication is a health priority among CKD patients

Finally, similar to other chronic conditions, CKD patients might experience mental illness CKD is associated with mental health issues such as cognitive impairment [117], depression [118], and even suicide attempt in later stages ESRD where dialysis

is required [119,120] Therefore, improving the mental health of CKD patients has been an important objective of CKD management programs

1.2 END-STAGE RENAL DISEASE

1.2.1 End-stage renal disease definition

End-stage renal disease (ESRD) is the most severe level of CKD ESRD is defined as

‘chronic kidney disease where kidney function is reduced to the extent that renal replacement therapy, i.e dialysis or kidney transplantation, is required to sustain life’

[26] In the US, the Centers for Medicare & Medicaid Services defined ESRD as ‘a

medical condition in which a person's kidneys cease functioning on a permanent basis leading to the need for a regular course of long-term dialysis or a kidney transplant

to maintain life’ [121] It is noted that stage 5 CKD and ESRD are not synonymous

21

This is because many patients with stage 5 CKD may live years without being symptomatic or needing RRTs, and conversely, some symptomatic patients with GFR>15 mL/min/1.73m2 may require RRTs Agarwal et al provided a more detail framework to define ESRD [122]

1.2.2 Epidemiology

The growth in prevalence and incidence of ESRD have been attributed to the expanding epidemic of hypertension and diabetes [123–126], aging population [127,128], and the improving patient survival predominantly as a result of innovative medical technology in dialysis and kidney transplantation [129–131] An estimated 2.6 million people received RRTs in 2010 worldwide [132] and the global prevalence of maintenance dialysis was 284 per million population (pmp), representing a 1.7 times increase compared to 1990 [133] The global incidence of ESRD rose notably from 44 pmp in 1990 to 93 pmp in 2010, representing a 2.5 times increase in universal healthcare countries and 1.8 times increase in countries with partial access to RRTs [133] However, the mismatch between the number of patients in need and the number

of patients accessing to RRTs is an issue Compared to the number of people in need

of RRTs, at least 2.3 million people might have died prematurely due to not being able

to access RRTs [132] More than 80% of patients receiving treatment for ESRD live

in countries with a large elderly population with access to universal health care [125] Worldwide use of RRTs was projected to rise to 5.4 million in 2030 with the highest growth in Asia [132] Figure 1.7 depicts the estimated annual incidence and prevalence

of ESRD in different countries in the world in 2013 As can be seen, the top five countries/territories with the highest annual incidence were Taiwan, the US, Mexico, Japan, and Shanghai The corresponding ranking for prevalence were Taiwan, Japan, the US, Portugal and Belgium [23]