Fibroblast Growth Factor-23 and Cardiac Magnetic Resonance Indices of

Myocardial Fibrosis in the Multi-Ethnic Study of Atherosclerosis

By

MOHAMED FAHER ALMAHMOUD, MD

A Thesis submitted to the Graduate Faculty of

WAKE FOREST UNIVERSITY GRADUATE SCHOOL OF ARTS AND SCIENCES

In Partial Fulfillment of the Requirements

for the Degree of

MASTER OF SCIENCE

Clinical Population and Translational Sciences

May 2016

Winston-Salem, North Carolina

Approved by:

David M. Herrington, MD, Mentor/Advisor

Beverly M. Snively, PhD

Jennifer Hawthorne Jordan, PhD, MS

i

ACKNOWLEDGMENTS

I am wholeheartedly grateful to Dr. David Herrington for his continued support,

guidance and mentorship during this fellowship and while preparing this thesis. I am very

thankful for every minute he spent teaching me by example what it means to be dedicated

and what it takes to be a successful clinical investigator. I am also grateful to Dr. Waqas

Qureshi who guided me to this excellent program and for being always available for any

questions. I am also grateful and appreciative of the time and support by Dr. Beverly

Snively who taught me basics of statistics. Thank you for reviewing every detail in this

work and for the thorough advices. I am also grateful to Dr. Jennifer Jordan for all her

support and encouragement. Thank you for answering my questions about the technical

basics of cardiac magnetic resonance imaging and T1 mapping. My strongest

appreciation and gratefulness to the CPTS program and the faculty who put every effort

to make sure I receive the best teaching I need to excell in all aspects of research. I would

also like to acknowledge Karen Blinson, Randi Sullivan, and Georgia Saylor for all their

assistance and support. My greatest gratitude goes to my mother Dr. Elham Tajeddin and

to my father Dr. Jamal Almahmoud. Thank you for inspiring me to seek the best

education and for supporting me during my research adventures. Thank you for being my

role models of dedication and hard work. And finally, I am most grateful to my wife, Dr.

Khulood Raslan, for her love and support at every step of my career.

ii

TABLE OF CONTENTS

LIST OF ILLUSTRATIONS AND TABLES……………………………………………V

LIST OF ABBREVIATIONS…………………………………………………………...Vii

ABSTRACT……………………………………………………………………………Viii

CHAPTER ONE: BACKGROUND………………………………………………………1

A. Introduction………………………………………………………………………..1

B. Definition of CKD………………………………………………………………...2

C. Cardiovascular disease in CKD population……………………………………….3

D. The pathophysiology of CVD in CKD……………………………………………4

E. Myocardial fibrosis in CKD………………………………………………………5

F. Fibroblast growth factor-23 in individuals with CKD……………………………9

G. The relationship between FGF-23 and all-cause and cardiovascular mortality….10

H. The role of FGF-23 in the pathogenesis of CVD in individuals with CKD……..10

I. Cardiac Magnetic Resonance indices of myocardial fibrosis……………………10

J. FGF-23 and CMR………………………………………………………………..15

K. Conceptual Model………………………………………………………………..15

L. Knowledge gap…………………………………………………………………..17

M. Aims and Hypotheses……………………………………………………………18

N. Methods………………………………………………………………………….19

O. References………………………………………………………………………..20

iii

CHAPTER TWO: MANUSCRIPT

A. KEYWORDS…………………………………………………………………….25

B. ABSTRACT……………………………………………………………………...26

Background………………………………………………………………………26

Methods…………………………………………………………………………..26

Results……………………………………………………………………………26

Conclusion……………………………………………………………………….27

C. INTRODUCTION……………………………………………………………….28

D. METHODS………………………………………………………………………29

Study population…………………………………………………………………29

Study procedure: CMR imaging…………………………………………………30

Measurement of serum FGF-23 concentrations………………………………….32

Statistical Analysis……………………………………………………………….32

E. RESULTS ……………………………………………………………………….34

F. DISCUSSION……………………………………………………………………35

G. LIMITATIONS…………………………………………………………………..38

H. CONCLUSION…………………………………………………………………..39

I. TABLES AND FIGURES……………………………………………………….40

J. SUPPLEMENTAL MATERIAL………………………………………………...44

K. REFERENCES…………………………………………………………………..45

iv

CAPTER 3: ANCILLARY ANLAYSES………………………………………………..49

A. ABSTRACT………………………….…………………………………………..50

B. INTRODUCTION……………………………………………………………….52

C. METHODS……………………….……………………………………………...52

Study population…………………………………………………………………53

Study procedures: CMR imaging………………………………………………...54

Measurement of serum FGF-23 concentrations………………………………….54

Ascertainment of heart failure events……………………………………………54

Statistical Analysis……………………………………………………………….55

D. RESULTS………………………………………………………………………..56

E. DISCUSSION……………………………………………………………………58

F. LIMITATIONS…………………………………………………………………..61

G. CONCLUSION…………………………………………………………………..61

H. TABLES AND FIGURES……………………………………………………….62

I. SUPPLEMENTAL MATERIAL………………………………………………...66

J. REFERENCES...………………………………………………………………...68

CURRICULUM VITAE…………………………………………………………………70

v

LIST OF ILLUSTRATIONS AND TABLES

CHAPTER ONE

Figure 1. Histological changes of the myocardial cells in CKD

Figure 2. Diffuse interstitial myocardial fibrosis in CKD (histologic evidence)

Figure 3. Patterns of myocardial fibrosis in CKD on cardiac magnetic resonance imaging

Figure 4. Regulation of FGF-23

Figure 5. Expected concentration range of FGF-23 as a function of eGFR

Figure 6. Conceptual model of the role of FGF-23 in the development of heart failure

CHAPTER TWO

Table I. Baseline characteristics by Fibroblast Growth Factor-23 quartiles

Table II. CMR indices of myocardial fibrosis by gender

Table III. Associations of FGF-23 with native myocardial T1 mapping, 12-min post-

contrast T1 mapping, and extracellular volume

Figure 1. Participant enrolment in the current study

Figure 2. Associations between serum FGF-23 quartiles and native myocardial T1

mapping, 12-min post-contrast myocardial T1 mapping and ECV

vi

SUPPLEMENTAL MATERIAL

Supplemental Table 1: Baseline participant’s characteristics by Fibroblast Growth

Factor-23 Quartile for the participants with measured ECV

CHAPTER THREE

Table I. Baseline characteristics by Fibroblast Growth Factor-23 quartiles

Table IIA. Association of FGF-23 and incident heart failure, HFrEF and HFpEF.

Table IIB. Association of FGF-23 and incident heart failure, HFrEF and HFpEF (Hazard

ratios per FGF-23 quartiles)

Figure I. Hazards ratios of the associations of FGF-23 quartiles with incident heart

failure, HFrEF and HFpEF.

Figure 2. Higher FGF-23 levels are associated with increased risk of incident heart

failure. Participants were stratified into quartiles of FGF-23. (Kaplan-Meier curve with

95% Hall-Wellner Bands and log-rank test). Notice the Y axis start from 90% survival, X

axis is the time to heart failure event (days).

Figure 3A. Higher FGF-23 levels are associated with increased risk of incident HFrEF.

Participants were stratified into quartiles of FGF-23. (Kaplan-Meier curve with 95%

Hall-Wellner Bands and log-rank test).

Figure 3B. Higher FGF-23 levels are associated with increased risk of incident HFpEF.

Participants were stratified into quartiles of FGF-23. (Kaplan-Meier curve with 95%

Hall-Wellner Bands and log-rank test).

Supplemental Table 1. Associations of FGF-23 and incident heart failure, HFrEF and

HFpEF stratified by gender (Hazard ratios per doubling of FGF-23).

Supplemental Table 2A. Associations of FGF-23 and incident heart failure, HFrEF and

HFpEF overall and stratified by gender without adjustment for LV mass.

Supplemental Table 2B. Associations of FGF-23 and incident heart failure, HFrEF and

HFpEF overall and stratified by gender with adjustment for LVH-ECG.

vii

LIST OF ABBREVIATIONS

CAD Coronary artery disease

CKD Chronic kidney disease

CKD-MBD Chronic kidney disease-Metabolic Bone disease

CMR Cardiovascular magnetic resonance

CV Cardiovascular

DE Delayed enhancement

ECV Extracellular volume

eGFR estimated Glomerular filtration rate

FGF-23 Fibroblast Growth Factor-23

HFpEF Heart failure with preserved ejection fraction

HFrEF Heart failure with reduced ejection fraction

LGE Late gadolinium enhancement

LV left ventricle

MESA Multi-Ethnic Study of Atherosclerosis

MOLLI Modified Look-Locker inversion recovery

MRI Magnetic resonance imaging

viii

ABSTRACT

In this thesis, we will evaluate the association of Fibroblast Growth Factor-23

(FGF-23) with myocardial fibrosis using cardiovascular magnetic resonance (CMR)

imaging techniques. Higher levels of Fibroblast Growth Factor-23 (FGF-23) have been

linked to increased risk of all-cause mortality, cardiovascular mortality, left ventricular

hypertrophy and heart failure. Histologic studies on rats suggest an association between

higher levels of FGF-23 with myocardial fibrosis. However, it is unknown whether such

relationship exists in humans.

We hypothesized that higher levels of FGF-23 would be associated with CMR indices of

myocardial fibrosis. We performed analyses of the association of FGF-23 with pre-and

post-contrast T1 mapping and extracellular volume (ECV) in participants from the Multi-

Ethnic Study of Atherosclerosis (MESA). The first chapter of this thesis will be a

literature review and will discuss the pathogenesis of myocardial fibrosis in individuals

with high levels of FGF-23 such as patients with chronic kidney disease (CKD). The

second chapter will be a manuscript of the associations of FGF-23 with CMR indices of

myocardial fibrosis. The third chapter is comprised of ancillary studies and discussion of

the association of serum FGF23 with incident heart failure with preserved ejection

fraction (HFpEF) and with reduced ejection fraction (HFrEF).

1

CHAPTER ONE: BACKGROUND

Introduction

Heart failure is common among patients with chronic kidney disease (CKD), and

is associated with increased mortality in this population.1 CKD patients have several

derangements of the metabolism of calcium and phosphate that are called metabolic bone

disease (CKD-MBD).2 These derangements lead ultimately to elevated blood levels of

fibroblast growth factor (FGF-23). Several studies showed the association between

elevation of FGF-23 and the development of heart failure.3,4

For example, Kestenbaum et

al. demonstrated an increased risk of incident heart failure with higher concentrations of

serum FGF-23 in a multi-ethnic middle aged population free of heart disease at baseline.3

The same relationship was shown to exist by Scilla et al. in a population with chronic

kidney disease stage 2-4.4 We aim to study the mechanism in which FGF-23 affects the

myocardium, specifically to confirm the findings of animal studies which link higher

levels of FGF-23 with increased myocardial fibrosis.5

Myocardial fibrosis is a common endpoint in a variety of cardiac diseases.6

Diffuse myocardial fibrosis occurs as a part of normal aging;7 however, this process is

accelerated in many diseases.8-10

This diffuse fibrosis is associated with worsening

ventricular function and increased ventricular stiffness and has independent predictive

value for major cardiac events and increased all-cause mortality.11-15

FGF-23 is a hormone secreted by osteoblasts/osteoclasts. It functions as a

phosphaturic factor in the renal tubules and a regulator of parathyroid hormone, thus

regulating calcium and phosphate homeostasis. Higher serum levels of FGF-23 were

2

shown to be associated with a variety of cardiovascular diseases. For example,

Kestenbaum et al. showed that higher serum FGF-23 concentrations are associated with

incident heart failure, left ventricular hypertrophy and coronary disease events.3

Recently, FGF-23 has been shown to be associated with myocardial fibrosis and left

ventricular hypertrophy in animal studies.16,17

The best available non-invasive method of examining myocardial fibrosis is

contrast-enhanced cardiac magnetic resonance (CMR) through T1 mapping. This

technique facilitates the quantification of diffuse myocardial fibrosis measured as the

extracellular volume fraction (ECV).

We propose to examine the association of serum FGF-23 with myocardial fibrosis in the

Multi-Ethnic Study of Atherosclerosis (MESA) using baseline serum FGF-23 and CMR

indices of myocardial fibrosis including ECV measurements and CMR-T1 mapping.

Definition of CKD

CKD is characterized by alterations in the kidney function for a minimum of 3

months. CKD is usually asymptomatic in the early stages.18

Symptoms generally are

related to complications of CKD that usually occur in the late stages. These

complications include cardiovascular disease (CVD),19

anemia,20

infectious

complications, neuropathy and abnormalities related to mineral bone metabolism.2

CKD was classified into five stages according to the National Kidney Foundation

criteria18

. This classification is based on both kidney function and structure. The

glomerular filtration rate (GFR) is used most commonly to assess function. For an

individual to be diagnosed with CKD, he/she must have at least two separate

3

measurements of GFR >90 days apart that are in the abnormal range (GFR<60

mL/min/1.73m2). Since many cases of kidney damage present with preserved GFR, other

markers of kidney damage can be used such as albuminuria, proteinuria, hematuria or

structural abnormalities. Thus, many patients with normal GFR with evidence of kidney

damage are considered to have CKD21

.

CKD has major prognostic implications22

, which include increased all-cause

mortality in patients with CKD compared with those without CKD. Most of the increased

mortality is attributed to increased cardiovascular events. The increased risk of

cardiovascular events is partly due to increased prevalence of traditional cardiovascular

risk factors such as hypertension, diabetes mellitus and also due to direct increased

cardiovascular risk from non-traditional risk factors such as anemia, malnutrition,

inflammation, retention of toxins and CKD-MBD.23,24

The advanced stages of CKD

which is classified as end stage renal disease (ESRD) is associated with even higher risk

of all-cause mortality compared with early stages of CKD.19

Cardiovascular disease in CKD population

CKD is increasingly recognized as an important risk factor for cardiovascular

(CV) mortality.25,26

Recent registry data from the US Renal Data System show that >40%

of ESRD deaths were due to CVD.1

The evidence of the relationship between renal dysfunction and adverse CV

events was first recognized in the dialysis population in whom incidence of CV mortality

is strikingly high. Approximately 50% of individuals with ESRD die from a CV

cause.27,28

In fact, the prevalence of CV disease is already high by the time patients reach

4

ESRD. For example, 40% of patients who have started dialysis have evidence of

coronary artery disease, and 85% of these patients have abnormal left ventricular

structure and function.29

Furthermore, left ventricular hypertrophy (LVH) was found to

be common in patients with CKD even before they progress to dialysis, and that the

prevalence of LVH correlates with the degree of renal functional impairment.30,31

This early link between CKD and CV events was further studied and was found to

follow graded correlation with decreasing level of glomerular filtration rate (GFR). In a

study by Go et al. showed a stepwise increase in the rate of mortality, CV events and

increased hospitalization. With the reference cohort with (GFR>60 mL/min), the adjusted

hazard ratio for death from any cause and any CV events increased to 1.2 and 1.4,

respectively, for GFR between 45 to 59 mL/min; 1.8 and 2.0 for GFR between 30 to 44

mL/min; 3.2 and 2.8 for GFR 15 to 29 mL/min; and 5.9 and 3.4 for GFR <15 mL/min.19

The pathophysiology of CVD in CKD

Even though patients with CKD have accelerated atherosclerosis and vascular

calcification, more patients die from arrhythmias (both atrial and ventricular), sudden

death, and congestive heart failure, than myocardial infarction1. Indeed, 50% of ESRD

patients suffer from heart failure (HF), thus forming the predominant cardiac abnormality

observed in CKD. It is a common perception that HF in CKD is secondary to vascular

comorbid conditions such as ischemic heart disease (IHD), hypertension and diabetes

mellitus, but it is becoming apparent that primary CKD is an independent risk factor for

incident HF even in non-diabetic and normotensive patients32

.

5

LVH in CKD is a pathologic process and unlike physiologic adaptations to

increased workload (e.g. “athletes heart”), is accompanied by fibrosis, which is attributed

to conditions related to the uremic milieu, including increased levels of parathyroid

hormone, endothelin, aldosterone, catecholamines, cardiotonic steroids and fibroblast

growth factors. In addition to fibrosis and cardiomyocyte hypertrophy, histological

changes of the heart in CKD also include myocyte apoptosis/necrosis resulting in

myocyte number reduction, and microvascular abnormalities such as arteriolar wall

thickening and decrease in the number of capillaries33-36

.

Myocardial fibrosis in CKD:

1- Histologic evidence

Diffuse myocardial interstitial widening and fibrosis is common in patients with

CKD.37

Several histological studies have shown increased myocardial fibrosis that is not

explained by other risk factors such as hypertension and/or diabetes. Experimental studies

have showen that interstitial fibrosis occur early after subtotal nephrectomy in the

absence of myocyte necrosis.38

Interstital deposition of collagen type I was associated

with swelling of interstitial cells – both cytoplasm and nuclei; in contrast, nuclear and

cytoplasmic volumes of endothelial cells were unchanged (Figure 1).

6

Figure 1: Histological changes of the myocardial cells in CKD

A. Myocardial interstitial cells of a control animal (electron micrograph, magnification 12,100:1). Cytoplasm without evidence of cell

activation. B. Activated interstitial cell of uremic rat three weeks after 5/6 nephrectomy (magnification 12,300:1). Note the increased

cytoplasm. Note the Abundant endoplasmic reticulum and Golgi complex and the numerous cytoskeletal filaments within the

cytoplasm.

Adapted from Mall et al. Kidney International, Vol.33 (1988), pp. 804-811

Furthermore, Aoki et al. demonstrated the pathologic characteristics of the heart of

dialysis patients with dilated cardiomyopathy which includes diffuse interstitial fibrosis

and severe myocyte hypertrophy with occasional disarray.36

In addition, the extent of the

left ventricular fibrosis was shown to be a strong predictor of cardiovascular mortality

(figure 2). The 3-year event-free cumulative survival rate for cardiac death was 42% in

dialysis patients with 30% or more fibrosis, whereas it was 82% in dialysis patients that

had less than 30% fibrosis (p=0.03). 36

7

Figure 2: Diffuse interstitial myocardial fibrosis in CKD (histologic evidence)

A 56-year-old male patient on dialysis for 7.1 years (A). Widespread fibrosis is present, and interstitial fibrosis surrounds each

myocyte (AZ stain, 400 xs). This patient died of ventricular arrythmia 1.1 years after biopsy. A 56-year-old male patient on dialysis

for 6.8 years (B). Only a little interstitial fibrosis present (AZ stain, 400 x). This patient had no cardiac events during 3.8 years of

follow-up.

Adapted from Aoki et al. Kidney International, Vol. 67 (2005), pp. 333-340

2- Imaging evidence:

By far the easiest way to study uremic changes of the myocardium in CKD patients is

echocardiography, which in studies, showed a graded relationship between the severity of

CKD and the prevalence and severity of LVH.25,30,31

Further and more precise

identification of CKD-related CVD can be demonstrated by CMR.

Mark et al. first used magnetic resonance imaging to discribed the patterns of

myocardial fibrosis in end stage renal disease patients. 5 Using late gadolinium

enhancement (LGE) as a way to detect fibrosis, they demonstrated two patterns. The first

pattern – subendocardial LGE, consistent with that described in myocardial infarction –

followed a primarily subendocardial distribution. The second pattern was characterized as

diffuse, less intense LGE reflecting diffuse myocardial fibrosis (figure 3). The latter

8

pattern was associated with greater LV mass compared with ESRD patients without LGE

(p<0.01).

Figure 3. Patterns of myocardial fibrosis in CKD on cardiac magnetic resonance imaging

(a) Short axis view of the left ventricle of hemodialysis patient demonstrating a diffuse area of gadolinium enhancement in the inferior

wall of the left ventricle (arrowed). Signal intensity of this area is 17.6 compared to the 6.9 for the LGE-negative area. (b) Short axis

view of the left ventricle of another hemodialysis patient demonstrating a diffuse area of gadolinium enhancement in the lateral wall of

the left ventricle. Signal intensity of the area of late gadolinium enhancement is 32.0 compared to 8.4 for the LGE-negative area. This

patient had normal coronary arteries at angiography performed as transplant assessment.

Adapted from Mark et al.; Kidney International (2006) 69, 1839-1845

Even though CMR is the best modality to detect myocardial fibrosis, its use is

currently restricted because of the risk of a fatal condition known as nephrogenic

systemic fibrosis with gadolinium use in patients with CKD. However, in a recent cross-

sectional study,39

43 patients with non-diabetic CKD stage 2-4 (mean GFR 50±22

ml/min/1.73m2) and 43 age- and gender-matched controls with no history of

cardiovascular disease underwent CMR imaging with T1 mapping. The patients with

CKD had higher mean native T1 times (986±37 ms) and mean extracellular volume

(28±4%) compared to controls (955±30ms and 25±3%, each p<0.05). This study

9

demonstrated that the increase in diffuse myocardial fibrosis occurred even in early

stages of CKD.39

Fibroblast growth factor in individuals with CKD

FGF-23 is a hormone secreted by osteoblasts and osteocytes and released into the

circulation.40

The effects of FGF-23 are mediated by FGF receptors (FGFRs), which are

located in the renal tubules (where FGF-23 induces urinary phosphate excretion)41

,

parathyroid gland (where FGF-23 regulates the production of parathormone [PTH]), and

the blood vessels (Figure 4). Many studies show that the coreceptor Klotho is mandatory

to induce FGF-23 specific signaling pathways.42-44

Furthermore, FGF-23 decreases dietary phosphate absorption through suppression

of circulating levels of 1, 25 dihydroxyvitamin D.45

In turn, vitamin D controls

production of FGF-23. It was shown that vitamin D3 stimulates FGF-23 generation. For

example, vitamin D receptor null mice showed undetectable FGF-23 levels.46

In addition, PTH is regulated by FGF-23, which suppresses both secretion and

gene expression of PTH. 47

However, patients with CKD have elevation of both PTH and

FGF-23, which can be explained by FGF-23 resistant status associated with uremia. It

was recently found that CKD is associated with Klotho deficiency which may explain

FGF-23 resistance.48

For example, in experimental models of CKD, there is a

downregulation of the FGF-23 receptor complex klotho-FGFR1 in the parathyroid,

resulting in resistance of the parathyroid to FGF-23.48-50

This resistance contributes to the

high levels of both PTH and FGF-23 in CKD. It was also shown that secondary

hyperparathyroidism is essential for the high levels of FGF-23 in CKD, as

10

parathyroidectomy both prevents and corrects the high FGF-23 levels of experimental

renal failure.51

Serum levels of FGF-23 increase with the decline of kidney function. These levels

can reach more than 200 times the normal levels in cases of ESRD (Figure 5).52

This rise

in FGF-23 levels occurs before changes in levels of serum phosphate, PTH, or 1, 25(OH)

2 vitamin D.

In summary, it has been observed

that FGF-23 levels increase with the decline

of kidney functioning through complex

mechanisms that include involvement of

bones, kidneys and the parathyroid gland.

Figure 4: Regulation of FGF-23

Adapted from: Larsson et al.

G Ital Nefrol 2014; 31(2)-ISSN 1724-5590

Figure 5: Expected concentration range of FGF-23 as a function of eGFR.

Adapted from: Larsson et al. G Ital Nefrol 2014; 31(2)-ISSN 1724-5590

11

The relationship between FGF-23 and all-cause and cardiovascular mortality

Increased FGF-23 levels in dialysis patients were associated with significantly

increased risk of mortality during the first year on dialysis; individuals with C-terminal

FGF-23 values above the median (1752 reference units (RU)/mL) were associated with

an odds ratio of 4.5-5.7 for mortality, compared to those with C-terminal FGF-23 <1089

RU/ml. 52

These results were confirmed even in predialysis CKD patients53

. Furthermore,

other studies also showed an association between higher levels of FGF-23 and increased

cardiovascular mortality and events.54

The role of FGF-23 in the pathogenesis of CVD in individuals with CKD

Recent studies have implied a role for FGF-23 in the pathophysiology of CKD-

mineral bone disease and in vascular calcification.55

It has been shown that FGF-23 levels

are elevated in CKD patients both on dialysis and on conservative treatment, which

showed association with increased mortality and left ventricular hypertrophy.16,52

Moreover, FGF-23 has been linked to increased arterial stiffness, endothelial dysfunction,

vascular calcification,56

and major cardiovascular events.4 Furthermore, a recent study in

animal models of renal failure showed reversal of uremic cardiomyopathy characterized

by left ventricular hypertrophy and fibrosis- with fibroblast growth factor blockade.57

Cardiac Magnetic Resonance indices of myocardial fibrosis

The myocardium is made of myocytes embedded in the myocardial interstitium,

which represents the scaffolding tissue. An expansion of this interstitium due to increased

collagen (focal and diffuse fibrosis) is the end result of many pathological processes such

as acute myocardial infarction, chronic myocardial infarction, and hypertrophic

12

cardiomyopathy.58

Expansion of the interstitium is also evident in many infiltrative

processes such as amyloidosis.59

The gold standard test to evaluate myocardial fibrosis is

histological assessment with invasive myocardial biopsy. However, the clinical utility of

this test is limited due to significant morbidity and sampling error.

Advancements in cardiac magnetic resonance (CMR) imaging have allowed for

the characterization of specific patterns of myocardial fibrosis. For example, late

gadolinium enhancement imaging (LGE) made it possible to identify focal areas of

myocardial infarction and enabled a detailed phenotypic characterization of many

nonischemic cardiomyopathies.60,61

However, the LGE technique is limited in detecting diffuse myocardial fibrosis

because it relies on qualitative detection of the signal intensity difference between fibrotic

and normal myocardial tissue. This limits the ability to detect homogeneously distributed

diffuse fibrosis, which is the main type in many forms of cardiomyopathy.62,63

The recent development of T1 mapping has allowed the quantitative assessment

of diffuse myocardial fibrosis. This technique measures the intrinsic magnetic resonance

parameter T1 of the myocardium and maps its spatial distribution (T1 mapping). The

measured differences in R1 (=1/T1) values pre- and post-gadolinium contrast allow the

quantification of the myocardial extracellular volume fraction (ECV), which has been

shown to be increased in the presence of diffuse myocardial fibrosis.64-66

T1 relaxation time of a tissue indicates how rapidly protons recover after a

radiofrequency pulse. Pre-contrast (native) T1 is affected by the water content and may

represent edema or diffuse myocardial fibrosis.67

Bull et al. studied whether non-contrast

13

CMR-T1 mapping sequence could identify myocardial fibrosis in patients with moderate

and severe aortic stenosis. They obtained biopsy samples for histologic assessment of

collagen volume fraction (CVF%) in 19 patients undergoing aortic valve replacement.

They showed a significant correlation between native T1 values and CVF% (r=0.65,

p=0.002). Higher T1 values were found in participants with severe aortic stenosis

compared to controls (972±33 ms vs. 944±16, p<0.05).67

However, this technique

embodies composite signal from both myocytes and interstitium, and it varies with the

magnetic resonance imaging field strength. Measurement of post-contrast T1 mapping

provides a value linked to the interstitium, but it is also influenced by gadolinium dose,

clearance rate, time post bolus, body composition, and hematocrit.

A newer, more accurate technique has been used recently to assess myocardial

fibrosis called ECV mapping. In this technique, ECV is measured after accounting for the

hematocrit level (ECV=100 X partition coefficient X [1-hematocrit]). The partition

coefficient is determined by the slope of the linear relationship of (1/T1myo vs. 1/T1blood).66

A novel CMR technique called Modified Look-Locker Inversion (MOLLI) recovery was

developed to allow the calculation of T1 mapping and ECV using a single breath-hold.68

This technique acquires a set of 11 source images in 17 heartbeats. It consists of 3

consecutively inversion recovery-prepared electrocardiography-synchronized Look-

Locker trains.

As mentioned above, ECV is increased in conditions characterized with diffuse

myocardial fibrosis. For example Broberg et al. found higher ECV in those with

congential heart disease compared to normal controls (31.9±5.8% versus 24.8±2%;

p<0.001). However, LGE did not account for the differences seen.64

Furthermore,

14

Ugander et al. showed the ability of ECV measurement of quantitatively characterizing

myocardial infarction (ECV infarct=51±8% vs ECV normal= 27±3%; p<0.001), atypical

diffuse fibrosis (ECV=37±6), and subtle myocardial abnormalities not clinically apparent

on LGE images.66

Recently in the Multi-Ethnic Study of Atherosclerosis, Liu et al.

showed increased ECV with older age in men. In addition, they showed a higher ECV in

females compared to males (28.1±2.8 vs. 25.8±2.9, p<0.001).7 Neilan et al.

69 studied the

distribution of ECV values in a healthy population and provided histological validation of

ECV measurements in mice. The normal range they found from studying 32 well-defined

healthy volunteers was (23 - 33%, mean 28±3%). In mice, they showed a significant

association between ECV and histological extent of fibrosis (r=0.94, p <0.001).

CMR T1 mapping was shown to correlate with myocardial fibrosis in many

different cardiac conditions such as acute myocardial infarction70

, chronic myocardial

infarction71

, acute myocarditis72

, aortic stenosis73

, hypertrophic cardiomyopathy74

,

congenital heart disease64

, idiopathic dilated cardiomyopathy, and infiltrative heart

diseae59

. These studies showed the ability of CMR-T1 mapping to detect diffuse

myocardial fibrosis, validated by histological evidence in several forms of

cardiomyopathy.

In summary, CMR imaging with T1 mapping and calculation of ECV is a novel

and validated non-invasive method to quantify the extent of diffuse myocardial fibrosis

even without the use of gadolinium. Furthermore, ECV measurement is superior to LGE

which can only give a qualitative assessment of focal areas of fibrosis.

15

FGF-23 and CMR:

Higher serum FGF-23 concentrations were associated with progressively greater

left ventricular mass and a greater prevalence of LVH. Kestenbaum et al. showed a linear

relationship between FGF-23 and left ventricular mass even after adjustment for

established LVH risk factors. For each 20-pg/mL higher serum FGF-23 concentration

there was an increase of 1.2 g in left ventricular mass (95% confidence interval, 0.2-2.2 g

greater) after full adjustment for common cardiovascular risk factors.3

Conceptual Model

Our understanding of the pathogenesis of heart failure and cardiovascular disease

among CKD patients is summarized by the conceptual model in (Figure 6). Abnormally

high levels of FGF-23 have several deleterious effects on the myocardium that ultimately

lead to heart failure. The model includes different mechanisms by which FGF-23

increases. Several changes occur with declining kidney function. While serum calcium,

and 1-25 (OH) vitamin D levels decline, PTH and phosphorus levels increase as part of

CKD-MBD. As a response to these derangements, FGF-23 increases and Klotho

decreases. Higher levels of FGF-23 are linked to a variety of clinical and pathological

conditions that lead to heart failure. Myocardial fibrosis is one of these conditions that

can be associated with higher levels of FGF-23. Furthermore, FGF-23 may lead to heart

failure with preserved ejection fraction (HFpEF) or diastolic heart failure as well as heart

failure with reduced ejection fraction (HFrEF) or systolic heart failure. Additionally,

Figure 6 lists possible confounders that may potentially lead to myocardial fibrosis as

16

well as the modifiers that can modify the relationship between FGF-23 levels and

myocardial fibrosis.

This conceptual model has several strengths, which include representation of the

current understanding of the pathophysiologic changes of CKD-MBD, and linking that to

the progression of heart failure throught FGF-23 based on a comprehensive literature

review. Also it acknowledges the complexity and the uncertainty of the current

understanding of this association between CKD and heart failure. Furthermore, it

considers the role of confounders and effect modifiers in the interactions between

antecedent, independent and dependent variables.

However, this conceptual model should also be considered with several

weaknesses, which include the inability to account for other mechanisms that interact

with FGF-23 levels, for example, dietary phosphate contents, the use of phosphate

binders or different vitamin D replacement preparations, aldosterone, estrogen, insulin

resistance and other unknown factors that may confound the proposed pathologic

mechanisms that link FGF-23 to heart failure.

17

Figure 6: Conceptual model of the role FGF-23 in the development of heart failure

Knowledge gap:

Though animal studies suggested a link between FGF-23 and myocardial

fibrosis,57

it is unknown whether this relationship is applicable to human populations.

Furthermore, the relationships between FGF-23 levels and CMR indices of myocardial

fibrosis have not been studied before. Additionally, though higher levels of FGF-23 were

shown to be associated with higher incidence of heart failure,3 it is yet to be determined

wither this relationship applies to both HFpEF and HFrEF. This work will be helpful

clinically and on a public health level by adding another tool of predicting myocardial

fibrosis and heart failure and thus allocating resources to prevention of the population at

risk at early stages.

18

Aims and Hypotheses

Primary Aim: To determine whether higher levels of fibroblast growth factor-23 (FGF-

23) are associated with indices of myocardial fibrosis detected by cardiac magnetic

resonance-T1 mapping in a multi-ethnic, middle aged community cohort free of heart

disease at baseline.

Secondary Aim: To determine whether FGF-23 is associated with heart failure with

preserved ejection fraction or with heart failure with reduced ejection fraction or both.

Predictor: baseline serum FGF-23

Outcomes: Extracellular volume fraction (ECV), pre-contrast T1 mapping, post-contrast

T1 mapping after 12 min, post-contrast T1 mapping after 25 min, ejection fraction,

incident HFpEF, incident HFrEF.

Primary Hypothesis: higher levels of FGF-23 are associated with higher levels of ECV

Secondary Hypothesis: FGF-23 levels are associated with both heart failure with

preserved ejection fraction and heart failure with reduced ejection fraction.

19

Methods:

We will address the first aim using a large cohort study of 1183 participants free

of cardiovascular diseae at baseline with FGF-23 measured at exam 1 (between 2000-

2002) and CMR T1 mapping performed at exam 5 (between 2010-2012). A subset of

these participants (n=588) had ECV calculated at exam 5. Participants with myocardial

scar (detected by visual assessment of any size using late gadolinium enhancement

imaging) were excluded. We will use multiple linear regression models to examine the

associations between FGF-23 and CMR indices of myocardial fibrosis (pre-contrast T1

mapping, 12-min and 25-min post-contrast T1 mapping and ECV %). ANOVA procedure

will be used to compare SAS driven least square means of CMR indices of fibrosis

between quartiles of FGF-23.

For the second aim we will use participants with baseline FGF-23 measured at

baseline exam 1 (n=6549) to examine the associations of FGF-23 with incident heart

failure, HFrEF and HFpEF. We will use Kaplan-Meier curves and log-rank procedure to

compare cumulative incidence of Heart failure, HFrEF and HFpEF by quartiles of FGF-

23. Cox proportional hazards regression models will be used to compute hazard ratios

using multivariate models.

20

References

1. Collins AJ, Foley RN, Herzog C, et al. US Renal Data System 2012 Annual Data Report. Am J Kidney Dis 2013;61:A7, e1-476. 2. Moe S, Drueke T, Cunningham J, et al. Definition, evaluation, and classification of renal osteodystrophy: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney international 2006;69:1945-53. 3. Kestenbaum B, Sachs MC, Hoofnagle AN, et al. Fibroblast growth factor-23 and cardiovascular disease in the general population: the Multi-Ethnic Study of Atherosclerosis. Circulation Heart failure 2014;7:409-17. 4. Scialla JJ, Xie H, Rahman M, et al. Fibroblast growth factor-23 and cardiovascular events in CKD. Journal of the American Society of Nephrology : JASN 2014;25:349-60. 5. Mark PB, Johnston N, Groenning BA, et al. Redefinition of uremic cardiomyopathy by contrast-enhanced cardiac magnetic resonance imaging. Kidney international 2006;69:1839-45. 6. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation 2000;101:2981-8. 7. Liu CY, Liu YC, Wu C, et al. Evaluation of age-related interstitial myocardial fibrosis with cardiac magnetic resonance contrast-enhanced T1 mapping: MESA (Multi-Ethnic Study of Atherosclerosis). Journal of the American College of Cardiology 2013;62:1280-7. 8. Schwartzkopff B, Frenzel H, Dieckerhoff J, et al. Morphometric investigation of human myocardium in arterial hypertension and valvular aortic stenosis. European heart journal 1992;13 Suppl D:17-23. 9. St John Sutton MG, Lie JT, Anderson KR, O'Brien PC, Frye RL. Histopathological specificity of hypertrophic obstructive cardiomyopathy. Myocardial fibre disarray and myocardial fibrosis. British heart journal 1980;44:433-43. 10. Tyralla K, Amann K. Morphology of the heart and arteries in renal failure. Kidney international Supplement 2003:S80-3. 11. Wong TC, Piehler KM, Kang IA, et al. Myocardial extracellular volume fraction quantified by cardiovascular magnetic resonance is increased in diabetes and associated with mortality and incident heart failure admission. European heart journal 2014;35:657-64. 12. Wong TC, Piehler K, Meier CG, et al. Association between extracellular matrix expansion quantified by cardiovascular magnetic resonance and short-term mortality. Circulation 2012;126:1206-16. 13. Gulati A, Jabbour A, Ismail TF, et al. Association of fibrosis with mortality and sudden cardiac death in patients with nonischemic dilated cardiomyopathy. Jama 2013;309:896-908. 14. Kwong RY, Korlakunta H. Diagnostic and prognostic value of cardiac magnetic resonance imaging in assessing myocardial viability. Topics in magnetic resonance imaging : TMRI 2008;19:15-24. 15. Conrad CH, Brooks WW, Hayes JA, Sen S, Robinson KG, Bing OH. Myocardial fibrosis and stiffness with hypertrophy and heart failure in the spontaneously hypertensive rat. Circulation 1995;91:161-70. 16. Faul C, Amaral AP, Oskouei B, et al. FGF23 induces left ventricular hypertrophy. The Journal of clinical investigation 2011;121:4393-408.

21

17. Xie J, Cha SK, An SW, Kuro OM, Birnbaumer L, Huang CL. Cardioprotection by Klotho through downregulation of TRPC6 channels in the mouse heart. Nature communications 2012;3:1238. 18. Levey AS, Eckardt KU, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney international 2005;67:2089-100. 19. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. The New England journal of medicine 2004;351:1296-305. 20. Astor BC, Muntner P, Levin A, Eustace JA, Coresh J. Association of kidney function with anemia: the Third National Health and Nutrition Examination Survey (1988-1994). Archives of internal medicine 2002;162:1401-8. 21. Gansevoort RT, Matsushita K, van der Velde M, et al. Lower estimated GFR and higher albuminuria are associated with adverse kidney outcomes. A collaborative meta-analysis of general and high-risk population cohorts. Kidney international 2011;80:93-104. 22. Chronic Kidney Disease Prognosis C, Matsushita K, van der Velde M, et al. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 2010;375:2073-81. 23. Longenecker JC, Coresh J, Powe NR, et al. Traditional cardiovascular disease risk factors in dialysis patients compared with the general population: the CHOICE Study. Journal of the American Society of Nephrology : JASN 2002;13:1918-27. 24. Fleischmann EH, Bower JD, Salahudeen AK. Are conventional cardiovascular risk factors predictive of two-year mortality in hemodialysis patients? Clinical nephrology 2001;56:221-30. 25. Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Hypertension 2003;42:1050-65. 26. Tonelli M, Wiebe N, Culleton B, et al. Chronic kidney disease and mortality risk: a systematic review. J Am Soc Nephrol 2006;17:2034-47. 27. Herzog CA, Ma JZ, Collins AJ. Poor long-term survival after acute myocardial infarction among patients on long-term dialysis. N Engl J Med 1998;339:799-805. 28. Foley RN, Parfrey PS, Sarnak MJ. Epidemiology of cardiovascular disease in chronic renal disease. Journal of the American Society of Nephrology : JASN 1998;9:S16-23. 29. Foley RN, Parfrey PS, Harnett JD, et al. Clinical and echocardiographic disease in patients starting end-stage renal disease therapy. Kidney Int 1995;47:186-92. 30. Tucker B, Fabbian F, Giles M, Thuraisingham RC, Raine AE, Baker LR. Left ventricular hypertrophy and ambulatory blood pressure monitoring in chronic renal failure. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 1997;12:724-8. 31. Levin A, Singer J, Thompson CR, Ross H, Lewis M. Prevalent left ventricular hypertrophy in the predialysis population: identifying opportunities for intervention. American journal of kidney diseases : the official journal of the National Kidney Foundation 1996;27:347-54. 32. Dhingra R, Gaziano JM, Djousse L. Chronic kidney disease and the risk of heart failure in men. Circulation Heart failure 2011;4:138-44. 33. Amann K, Wanner C, Ritz E. Cross-talk between the kidney and the cardiovascular system. J Am Soc Nephrol 2006;17:2112-9.

22

34. Kennedy DJ, Vetteth S, Periyasamy SM, et al. Central role for the cardiotonic steroid marinobufagenin in the pathogenesis of experimental uremic cardiomyopathy. Hypertension 2006;47:488-95. 35. London GM. Left ventricular alterations and end-stage renal disease. Nephrol Dial Transplant 2002;17 Suppl 1:29-36. 36. Aoki J, Ikari Y, Nakajima H, et al. Clinical and pathologic characteristics of dilated cardiomyopathy in hemodialysis patients. Kidney international 2005;67:333-40. 37. Mall G, Huther W, Schneider J, Lundin P, Ritz E. Diffuse intermyocardiocytic fibrosis in uraemic patients. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 1990;5:39-44. 38. Mall G, Rambausek M, Neumeister A, Kollmar S, Vetterlein F, Ritz E. Myocardial interstitial fibrosis in experimental uremia--implications for cardiac compliance. Kidney international 1988;33:804-11. 39. Edwards NC, Moody WE, Yuan M, et al. Diffuse interstitial fibrosis and myocardial dysfunction in early chronic kidney disease. The American journal of cardiology 2015;115:1311-7. 40. Larsson T, Marsell R, Schipani E, et al. Transgenic mice expressing fibroblast growth factor 23 under the control of the alpha1(I) collagen promoter exhibit growth retardation, osteomalacia, and disturbed phosphate homeostasis. Endocrinology 2004;145:3087-94. 41. Gattineni J, Bates C, Twombley K, et al. FGF23 decreases renal NaPi-2a and NaPi-2c expression and induces hypophosphatemia in vivo predominantly via FGF receptor 1. American journal of physiology Renal physiology 2009;297:F282-91. 42. Dailey L, Ambrosetti D, Mansukhani A, Basilico C. Mechanisms underlying differential responses to FGF signaling. Cytokine & growth factor reviews 2005;16:233-47. 43. Kurosu H, Ogawa Y, Miyoshi M, et al. Regulation of fibroblast growth factor-23 signaling by klotho. The Journal of biological chemistry 2006;281:6120-3. 44. Razzaque MS, Lanske B. The emerging role of the fibroblast growth factor-23-klotho axis in renal regulation of phosphate homeostasis. The Journal of endocrinology 2007;194:1-10. 45. Shimada T, Hasegawa H, Yamazaki Y, et al. FGF-23 is a potent regulator of vitamin D metabolism and phosphate homeostasis. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2004;19:429-35. 46. Shimada T, Yamazaki Y, Takahashi M, et al. Vitamin D receptor-independent FGF23 actions in regulating phosphate and vitamin D metabolism. American journal of physiology Renal physiology 2005;289:F1088-95. 47. Ben-Dov IZ, Galitzer H, Lavi-Moshayoff V, et al. The parathyroid is a target organ for FGF23 in rats. The Journal of clinical investigation 2007;117:4003-8. 48. Canalejo R, Canalejo A, Martinez-Moreno JM, et al. FGF23 fails to inhibit uremic parathyroid glands. Journal of the American Society of Nephrology : JASN 2010;21:1125-35. 49. Galitzer H, Ben-Dov IZ, Silver J, Naveh-Many T. Parathyroid cell resistance to fibroblast growth factor 23 in secondary hyperparathyroidism of chronic kidney disease. Kidney Int 2010;77:211-8. 50. Komaba H, Goto S, Fujii H, et al. Depressed expression of Klotho and FGF receptor 1 in hyperplastic parathyroid glands from uremic patients. Kidney Int 2010;77:232-8. 51. Lavi-Moshayoff V, Wasserman G, Meir T, Silver J, Naveh-Many T. PTH increases FGF23 gene expression and mediates the high-FGF23 levels of experimental kidney failure: a bone parathyroid feedback loop. Am J Physiol Renal Physiol 2010;299:F882-9.

23

52. Gutierrez OM, Mannstadt M, Isakova T, et al. Fibroblast growth factor 23 and mortality among patients undergoing hemodialysis. The New England journal of medicine 2008;359:584-92. 53. Isakova T, Xie H, Yang W, et al. Fibroblast growth factor 23 and risks of mortality and end-stage renal disease in patients with chronic kidney disease. Jama 2011;305:2432-9. 54. Kendrick J, Cheung AK, Kaufman JS, et al. FGF-23 associates with death, cardiovascular events, and initiation of chronic dialysis. Journal of the American Society of Nephrology : JASN 2011;22:1913-22. 55. Nitta K, Nagano N, Tsuchiya K. Fibroblast growth factor 23/klotho axis in chronic kidney disease. Nephron Clinical practice 2014;128:1-10. 56. Desjardins L, Liabeuf S, Renard C, et al. FGF23 is independently associated with vascular calcification but not bone mineral density in patients at various CKD stages. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 2012;23:2017-25. 57. Di Marco GS, Reuter S, Kentrup D, et al. Treatment of established left ventricular hypertrophy with fibroblast growth factor receptor blockade in an animal model of CKD. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 2014;29:2028-35. 58. Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. Journal of the American College of Cardiology 2011;57:891-903. 59. Mongeon FP, Jerosch-Herold M, Coelho-Filho OR, Blankstein R, Falk RH, Kwong RY. Quantification of extracellular matrix expansion by CMR in infiltrative heart disease. JACC Cardiovascular imaging 2012;5:897-907. 60. Kim RJ, Chen EL, Lima JA, Judd RM. Myocardial Gd-DTPA kinetics determine MRI contrast enhancement and reflect the extent and severity of myocardial injury after acute reperfused infarction. Circulation 1996;94:3318-26. 61. Karamitsos TD, Francis JM, Myerson S, Selvanayagam JB, Neubauer S. The role of cardiovascular magnetic resonance imaging in heart failure. Journal of the American College of Cardiology 2009;54:1407-24. 62. de Leeuw N, Ruiter DJ, Balk AH, de Jonge N, Melchers WJ, Galama JM. Histopathologic findings in explanted heart tissue from patients with end-stage idiopathic dilated cardiomyopathy. Transplant international : official journal of the European Society for Organ Transplantation 2001;14:299-306. 63. Heymans S, Schroen B, Vermeersch P, et al. Increased cardiac expression of tissue inhibitor of metalloproteinase-1 and tissue inhibitor of metalloproteinase-2 is related to cardiac fibrosis and dysfunction in the chronic pressure-overloaded human heart. Circulation 2005;112:1136-44. 64. Broberg CS, Chugh SS, Conklin C, Sahn DJ, Jerosch-Herold M. Quantification of diffuse myocardial fibrosis and its association with myocardial dysfunction in congenital heart disease. Circulation Cardiovascular imaging 2010;3:727-34. 65. Jerosch-Herold M, Sheridan DC, Kushner JD, et al. Cardiac magnetic resonance imaging of myocardial contrast uptake and blood flow in patients affected with idiopathic or familial dilated cardiomyopathy. Am J Physiol Heart Circ Physiol 2008;295:H1234-H42. 66. Ugander M, Oki AJ, Hsu LY, et al. Extracellular volume imaging by magnetic resonance imaging provides insights into overt and sub-clinical myocardial pathology. European heart journal 2012;33:1268-78.

24

67. Bull S, White SK, Piechnik SK, et al. Human non-contrast T1 values and correlation with histology in diffuse fibrosis. Heart 2013;99:932-7. 68. Messroghli DR, Radjenovic A, Kozerke S, Higgins DM, Sivananthan MU, Ridgway JP. Modified Look-Locker inversion recovery (MOLLI) for high-resolution T1 mapping of the heart. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2004;52:141-6. 69. Neilan TG, Coelho-Filho OR, Shah RV, et al. Myocardial extracellular volume fraction from T1 measurements in healthy volunteers and mice: relationship to aging and cardiac dimensions. JACC Cardiovascular imaging 2013;6:672-83. 70. Dall'Armellina E, Piechnik SK, Ferreira VM, et al. Cardiovascular magnetic resonance by non contrast T1-mapping allows assessment of severity of injury in acute myocardial infarction. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2012;14:15. 71. Messroghli DR, Walters K, Plein S, et al. Myocardial T1 mapping: application to patients with acute and chronic myocardial infarction. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2007;58:34-40. 72. Ferreira VM, Piechnik SK, Dall'Armellina E, et al. Non-contrast T1-mapping detects acute myocardial edema with high diagnostic accuracy: a comparison to T2-weighted cardiovascular magnetic resonance. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2012;14:42. 73. Flett AS, Sado DM, Quarta G, et al. Diffuse myocardial fibrosis in severe aortic stenosis: an equilibrium contrast cardiovascular magnetic resonance study. European heart journal cardiovascular Imaging 2012;13:819-26. 74. Flett AS, Hayward MP, Ashworth MT, et al. Equilibrium contrast cardiovascular magnetic resonance for the measurement of diffuse myocardial fibrosis: preliminary validation in humans. Circulation 2010;122:138-44.

25

CHAPTER TWO: MANUSCRIPT

Fibroblast Growth Factor-23 and Cardiac-MRI Indices of

Myocardial Fibrosis in the Multi-Ethnic Study of

Atherosclerosis

Short Title: FGF-23 and CMR indices of myocardial fibrosis in MESA.

Mohamed Faher Almahmoud, MD1, Jennifer H Jordan, PhD, MS

1, Bryan Kestenbaum,

MD2, Ronit Katz, PhD

2, Bharath Amble Venkatesh, PhD

3, João A. C. Lima, MD

3,4, Chia-

Ying Liu, PHD5, Erin D. Michos, MD, M.H.S

4, David M Herrington, MD, M.H.S

1

1Department of Internal Medicine, Section on Cardiovascular medicine, Wake Forest School of

Medicine, Winston-Salem, NC, USA

2Kidney Research Institute, Department of Medicine, Division of Nephrology, University of

Washington, Seattle, WA, USA

3Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, USA

4Division of Cardiology, Johns Hopkins University School of Medicine, Baltimore, MD, USA

5Radiology and Imaging Sciences, National Institutes of Health (NIH), Bethesda, MD, USA

Correspondence:

Mohamed Faher Almahmoud MD, Department of Internal Medicine, Section on Cardiology

Wake Forest School of Medicine

Medical Center Boulevard, Winston-Salem, NC

Tel: (336)716.2255, Fax: (336)716.3202

Email: malmahmo@wakehealth.edu

26

Abstract

Background: Fibroblast growth factor-23 (FGF-23) is associated with myocardial

fibrosis in animal studies. We examined the associations between FGF-23 and Cardiac

Magnetic Resonance (CMR) indices of myocardial fibrosis using T1 mapping.

Methods: A total of 6547 study participants free of heart disease at baseline, had serum

FGF-23 concentrations measured between 2000-2002. We included 1183 participants

(49% Male; age 58±9 at baseline) with eGFR>45 mL/min/1.73m2 who underwent CMR

between 2010-2012 and had no evidence of focal scars on late gadolinium enhancement.

ANOVA and multivariable linear regression models were performed to determine the

relationships between baseline log2 transformed FGF-23 levels and follow up myocardial

T1 mapping (pre-contrast, 12- and 25-min post-contrast), and extracellular volume

fraction (ECV %) (n=588). Models were adjusted for cardiovascular risk factors, left

ventricular mass and stratified by gender.

Results: Overall, we found no significant relationships between FGF-23 and CMR

indices of myocardial fibrosis. However, in gender stratified, a doubling of FGF-23 was

associated with an absolute 1.2% higher ECV (95% CI: 0.33, 2.05%; p=0.007) and a

near significant 7.1(ms) higher pre-contrast myocardial T1 mapping (95% CI: -0.1, 14.4;

p=0.053) in men. Overall there was a trend for a positive relationship between FGF-23

quartiles and pre-contrast myocardial T1 mapping (p-trend =0.02). After stratification,

men in the highest quartile (FGF-23 >46.1 pg/ml) had significantly higher pre-contrast

myocardial T1 mapping and ECV values (973±3 (ms), 26.3±0.3%, respectively)

27

compared to those in the lowest quartile (FGF-23 <31 pg/ml) (961±4 (ms), 24.7±0.4%,

respectively) (p value=0.002 and 0.01 respectively). In contrast, we found no significant

association between FGF-23 and CMR indices of myocardial fibrosis in women.

Conclusion: In men FGF-23 is positively associated with CMR indices of myocardial

fibrosis. This mechanism may explain the increased risk for heart failure associated with

higher levels of FGF-23.

28

Introduction

Myocardial fibrosis is a common endpoint in a variety of cardiac diseases.1

Diffuse myocardial fibrosis occurs as a part of normal aging;2 however, this process is

accelerated in many diseases.3-5

This diffuse fibrosis is associated with worsening

ventricular function and increased ventricular stiffness, and has independent predictive

value for major cardiac events and increased all-cause mortality.6-10

The pathogenesis of

myocardial fibrosis, which leads to significant increase in collagen deposition in the

myocardium, is yet not fully understood.

Recently, fibroblast growth factor-23 (FGF-23) and its coreceptor Klotho have

been shown to be associated with myocardial fibrosis and left ventricular hypertrophy in

animal studies.11,12

FGF-23 is a hormone secreted by osteoblasts/osteoclasts. It functions

as a phosphaturic factor in the renal tubules and a regulator of parathyroid hormone, thus

regulating calcium and phosphate homeostasis. Higher serum levels of FGF-23 were

shown to be associated with variety of cardiovascular cardiac diseases.13,14

For example,

Kestenbaum et al showed that higher serum FGF-23 concentrations are associated with

incident heart failure, left ventricular hypertrophy and coronary disease events in a

population free of heart disease at baseline13

. However, it is unknown whether this

increase in incident heart failure and left ventricular hypertrophy is associated with

increased myocardial fibrosis.

The best available non-invasive method of examining myocardial fibrosis is

contrast-enhanced cardiac magnetic resonance (CMR). The use of late gadolinium

enhancement imaging (LGE) enabled the identification of focal myocardial fibrosis.15

However, this technique is limited in detecting diffuse myocardial fibrosis because it

29

depends on the signal intensity difference between fibrotic and nonfibrotic tissues which

is reduced in homogeneously distributed fibrosis.16

Recent development of T1 mapping allowed the quantitative assessment of

diffuse myocardial fibrosis through the measurement of the intrinsic magnetic resonance

relaxation parameter T1 of the myocardium.17

T1 mapping time (in milliseconds)

represents protons recovery time after radiofrequency pulse. Longer pre-contrast (native)

T1 mapping and shorter post-contrast T1 mapping were shown to be associated with

diffuse myocardial fibrosis.18,19

A newer more accurate technique allows quantification of the extracellular

volume fraction (ECV) by measuring the ratio of change in myocardial T1 relative to

blood-pool T1 pre and post-contrast after correcting for hematocrit.20

ECV is increased in

conditions characterized with diffuse myocardial fibrosis.21,22

In a large multi-ethnic cohort of men and women, we hypothesized that higher levels of

FGF-23 would be associated with indices of myocardial fibrosis.

Methods

Study population

The Multi-Ethnic Study of Atherosclerosis (MESA) is a prospective cohort study

investigating risk factors and progression of subclinical cardiovascular disease among

6814 community-living individuals. 23

Between 2000 and 2002, the MESA recruited

participants free of cardiovascular disease aged 45 to 84 years from six sites from United

States, Baltimore, MD; Chicago, IL; St. Paul, MN; Forsyth County, NC; New York, NY;

and Los Angeles, CA. The MESA recruited a final population that was 38% white, 28%

black, 22% Hispanic, and 12% Chinese-Americans. The study excluded any participants

30

with self-reported history of myocardial infarction, angina, stroke, transient ischemic

attack, heart failure, atrial fibrillation, nitroglycerin use, angioplasty, coronary artery

bypass grafting, valve replacement, pacemaker or defibrillator, or any surgery on the

heart or arteries. All participants gave informed consent, and institutional review board

approval was obtained for each site.

We evaluated all participants who had FGF-23 measurement at baseline and

subsequently had cardiac magnetic resonance (CMR) with T1 mapping (n=1183) and

ECV% measurement (n=588) on the 5th

follow-up exam (between 2010-2012). Refer to

(Figure 1) for explanation of participants’ enrollment in this study.

Study procedures: CMR imaging.

MESA participants without contraindications underwent CMR examinations by

using 1.5-T scanners (Avanto and Espree, Siemens Medical Systems, Eelangen,

Germany) with a 6-channel anterior phased array torso coil and corresponding posterior

coil elements. Left ventricular (LV) function, dimensions, and myocardial mass were

assessed by a cine steady-state free precession sequence. Twelve short axis slices, one 4-

chamber view, and one 2-chamber view were acquired. Participants undergoing CMR

scans were screened for gadolinium eligibility. The inclusion criteria were the estimated

glomerular filtration rates (eGFRs). Participants with an eGFR ≥45 ml/min (60 ml/min

for the site at Northwestern University) and with no history of allergic reaction to contrast

agents were qualified to receive gadolinium. Late gadolinium enhancement images were

obtained 15 min after an intravenous bolus injection of gadolinium-diethylene triamine

pentaacetic acid (0.15 mmol/kg [Magnevist, Bayer Healthcare Pharmaceuticals,

Montville, New Jersey]) to identify myocardial fibrosis. Twelve short-axis slices, 1

31

horizontal long axis, and 1 vertical long axis at the same positions as the LV function

cine images were acquired.

For evaluation of diffuse fibrosis, 1 short axis MOLLI image at the mid-slice

position was acquired pre-contrast (native), then repeated at 12 and 25 min post-contrast

administration. 12-and 25 min post-contrast T1 maps are used to draw the inversion

recovery curve. The timing was chosen to be comparable with previous studies and also

to accommodate the design of the entire CMR protocol.24-26

The MOLLI sequence

acquired a set of 11 source images in 17 heartbeats. It consisted of 3 consecutively

inversion recovery-prepared electrocardiography-synchronized Look-Locker trains. Each

of the 3 trains began with an inversion pulse at specific inversion time (T1= 100, 200, and

350 ms), after which multiple single shot, steady-state free precession images were

acquired in consecutive heartbeats. All images were acquired with the same trigger delay

time in end diastole. The exact scanning parameters were as follows: flip angle =35o;

repetition time=2.2 ms; echo time=1.1 ms; field of view=360 X 360 mm; matrix=192 X

183; slice thickness = 8 mm; generalized autocalibrating partially parallel acquisitions

factor=2.

We constructed T1 maps offline by using MASS research software (Department

of Radiology, Leiden University Medical Center, Leiden, The Netherland). We

performed a 3-parameter curve firt of the MOLLI source images according to the

Levenberg-Marquardt algorithm to calculate T1 values for each pixel automatically.

Then, we selected a region of interest around the core of myocardium to exclude the

blood pool and epicardial fat and calculate the myocardial T1 time for each subject. T1

maps were calculated pre-contrast (native) and 12-min, 25 min post-contrast. The

32

extracellular volume fraction (ECV [%]) was determined using the following equation

(ECV=100 X partition coefficient X [1- hematocrit]). To determine the partition

coefficient, we used the slope of the linear relationship of (1/T1 myo vs. 1/T1blood) at three

time points as described in previously published studies. 2,26-28

Because ECV calculation

depends on both T1 maps and hematocrit level at the time of CMR performance, only a

subset of the participants (n=608) with available hematocrit data at the time of CMR

examination had ECV calculated. Of them 588 with available FGF-23 data were included

in our analysis (Figure 1).

Measurement of Serum FGF-23 Concentrations

The University of Vermont Laboratory for Clinical Biochemistry stored fasting

blood and urine samples collected by MESA study personnel using established

methods.29

These samples were shipped on dry ice to the University of Washington

where serum FGF-23 concentrations were measured using the Kainos Immunoassay30

which detects the full-length biologically intact FGF-23 molecule via mid-molecule and

distal epitopes. Standarized FGF-23 controls within each run were used to monitor

quality control. The coefficient of variation across 81 plates was 6.7% and 12.4% for

high and low control samples respectively.13

Statistical Analysis

We examined the distribution of the variables and presented continuous variables

as mean ± SD and categorical variables as frequencies and percentages. Participant

characteristics were presented by FGF-23 quartiles. We used multiple linear regression

models to estimate associations of baseline serum FGF-23 concentrations and follow-up

CMR variables of myocardial fibrosis (native, 12- and 25-min post-contrast myocardial

33

T1 mapping in milliseconds (ms), and ECV %). We repeated multiple linear regression

models after dividing subjects into quartiles based on serum FGF-23 concentrations. Due

to skewed distributions, FGF-23 was explored as a continuous predictor variable after log

base 2 transformations so the parameter coefficient can be interpreted as “per doubling of

FGF-23.” Because of previous observations of gender differences in cardiac remodeling

and myocardial fibrosis,2,31

we performed the analyses overall and stratified by gender.

We performed linear regressions with multivariate adjustments for demographic variables

and risk factors. The initial model for all analyses was unadjusted. Model 2 was adjusted

for age, race, study site, highest education level completed (high school or less, some

college but no degree, college degree or more), height and weight. Model 3 was further

adjusted for systolic blood pressure (continuous), antihypertensive medications (yes

versus no), LV mass, heart rate, LDL, HDL, diabetes mellitus status (yes versus no), C-

reactive protein (CRP), smoking (current or former versus nonsmoker), urine albumin-

creatinine ratio, and eGFRCKD-Epi. P<0.05 was considered significant for all analyses

including interaction terms. All statistical analyses were performed with SAS software,

version 9.3 (SAS Institute, Cary, NC).

Because we had 2 different datasets, one with native and post-contrast T1

mapping measurements (n=1183) and a subset with ECV measurements (n=588), we had

mathematically but not clinically different cutoffs of FGF-23 quartiles. We used quartiles

cutoffs from the main dataset (n=1183) for the analyses. However, when repeating the

analysis of the association of FGF-23 with ECV based on FGF-23 quartiles specific for

the ECV dataset (n=588), we found similar results. Supplemental table 2 shows the

participant’s baseline characteristics by FGF-23 quartiles specific to the ECV dataset.

34

Results

A total of 1183 (49% men) participants with baseline FGF-23 and follow up T1

mapping free of focal scars on late gadolinium enhancement were included in the study.

Of them 588 (47% men) participants had calculated ECV values (Figure 1). Baseline

characteristics of the participants (n=1183) were presented according to quartiles of

serum FGF-23 levels (Table 1). Overall, 52.4% of the participants were white, 22%

blacks, 13.8% hispanics and 11.8% Chinese American. For the subset with calculated

ECV, 58% were white, 28% blacks and 14% hispanics. Mean eGFR was 83.8 mL/min

per 1.73 m2. There was a gradual decline in mean eGFR with increasing serum FGF-23

quartiles (P <0.001). For doubling of FGF-23, eGFR decreased by 5.67 ml/min per 1.73

m2 (95% CI 3.88, 7.47. p value <0.001). Participants in the 4

th quartile had higher body

mass index (BMI). The participant’s characteristics of the ECV dataset were presented in

supplemental table 2.

Women had significantly higher native myocardial T1 mapping and ECV and

lower 12-min and 25-min post-contrast myocardial T1 mapping compared to men (Table

2). The interaction between gender and log2FGF-23, as measured by the regression

coefficient for men and women, was near significant for native T1 mapping, and

nonsignificant for 12-min post-contrast T1 mapping and ECV (p=0.07, p=0.4 and p=0.1,

respectively) in the fully adjusted model.

Overall, there was no significant relationship between serum FGF-23 and CMR

indices of myocardial fibrosis (Table 3). However, after stratification by gender, a

doubling of FGF-23 was associated with an absolute 1.2% higher ECV (95% CI: 0.33,

35

2.05%; p=0.007) and a near significant increase in native myocardial T1 mapping by

7.1(ms) (95% CI: -0.1, 14.4 (ms); p=0.053) in men only.

When comparing adjusted mean values of CMR indices of fibrosis by quartiles of

serum FGF-23 (Figure 2), overall there was a trend for higher native myocardial T1

mapping (p-trend=0.02). After stratification by gender, men had increasing native

myocardial T1 mapping and ECV values with higher serum FGF-23 quartiles. The

adjusted mean difference in native myocardial T1 mapping for men in the 4th

quartile

compared to the 1st quartile was 12 (ms) (95% CI: 3, 21 (ms), p-trend=0.002) and an

absolute difference of 1.55% in ECV (95% CI: 0.6, 2.5%; p-trend =0.01). No statistically

significant relationship between serum FGF-23 concentration and either native

myocardial T1 mapping or ECV was seen in women.

We found no significant relationship between FGF-23 and 12-min, 25-min post-

contrast myocardial T1 mapping. We found high correlation between the two post-

contrast myocardial T1 mapping variables [r2] =0.84, thus only 12-min post-contrast

myocardial T1 mapping was presented in table (3) and figure (2).

Discussion

In the Multi-Ethnic Study of Atherosclerosis, we evaluated the relationship

between serum FGF-23 concentrations and myocardial tissue composition by using CMR

T1 mapping. Overall, we found no significant relationship between serum FGF-23

concentrations and indices of myocardial fibrosis. However, after stratification by gender,

we found a significantly positive relationship between serum FGF-23 concentration and

both native myocardial T1 mapping and extracellular volume in men. Additionally, we

36

found a significant trend between quartiles of FGF-23 and native myocardial T1 mapping

overall and in men. These findings can be interpreted as an increase in the collagen

composition of the myocardium and thus increased diffuse myocardial fibrosis. However,

we did not find statistically significant associations between serum FGF-23 concentration

and CMR indices of myocardial fibrosis in women.

Myocardial fibrosis is a pathological process observed as a response to

myocardial injury. The gold standard method to detect diffuse myocardial fibrosis is

endomyocardial biopsy.32

However, this method is invasive and is subject to sampling

error. CMR T1 mapping has been shown to provide a non-invasive validated

quantification of cardiac tissue composition and fibrosis.16

While focal fibrosis usually

follows myocardial infarction,33

diffuse myocardial fibrosis has been detected in non-

ischemic cardiomyopathy,34

hypertrophic cardiomyopathy,35

heart failure,24

diabetes,36

atrial fibrillation,37

aging,2 and infiltrative cardiomyopathies.

38 It is commonly associated

with LV hypertrophy,39

abnormal cardiac remodeling,40

and worsening ventricular

systolic and diastolic function.41,42

FGF-23 regulates calcium and phosphorus hemostasis through binding to FGF

receptors and klotho, its coreceptor in the kidney and parathyroid glands.43,44

FGF-23

concentrations increase and soluble klotho concentrations decrease in chronic kidney

disease with progressive decline of phosphorus excretion.45

Previous animal studies

suggested a causal role for FGF-23 in the development of cardiovascular disease. In

mice, intravenous injection of FGF-23 induced pathological hypertrophy of isolated

cardiomyocytes by klotho-independent mechanism.11

Administration of FGF receptors

blocker to a rat model of CKD attenuated the severity of LVH without reducing elevated

37

blood pressure.11

Furthermore, higher levels of FGF-23 concentrations were associated

to increased myocardial fibrosis and remodeling in presence of moderate or severe klotho

deficiency.46

This pathologic elevation in FGF-23 provides a key mechanism of the abnormally

high prevalence of LV hypertrophy,47

heart failure,48

and myocardial fibrosis,49

as well as

the increased cardiovascular mortality noted in patients with CKD.50,51

This is important

because FGF-23 levels start increasing at early stages of CKD (eGFR <75). In fact, it was

found that CKD was associated with increased myocardial fibrosis even at early stages.52

Our findings of differences in CMR indices of myocardial fibrosis between men

and women are consistent with previous study of age-related effect on myocardial

fibrosis by Liu et al,2 which demonstrated significantly higher ECV in women, and more

pronounced increase in ECV with age in men. Our findings of different response of the

myocardium to higher levels of FGF-23 are also consistent with previous studies of

gender differences in myocardial response to different stimuli.31

For example, while

women tend to preserve their myocardial mass, myocyte number and volume, men lose

myocardial mass and myocyte number, but increase in myocyte cell volume.53

Furthermore, in a post-mortem study, Mallat et al. showed that the apoptotic index was

found to be 3-fold higher in men compared to women free of cardiovascular disease.54

Compared to men, women have greater degree of LV hypertrophy and preserved LV

function in response to pressure overload.55

Similar findings were seen as a response to

volume overload. For example, Gardner et al. demonstrated that female rats developed

concentric hypertrophy with no impairment of cardiac function and no changes in

myocardial compliance after 8 weeks of infra-renal aortocaval fistula creation. In contrast

38

male rats developed a significant fistula-related dilation and increase ventricular

compliance.56

Thus, women may be protected from fibrosis.

We found no significant relationship between FGF-23 and 12-min, 25-min post-

contrast myocardial T1 mapping. However, measurement of post-contrast myocardial T1

mapping provides a value linked to the interstitium, but it is also influenced by

gadolinium dose, clearance rate, time post bolus, body composition, and hematocrit.

Because of these limitations, native T1 mapping and ECV measurement are preferred

over post-contrast T1 mapping for the detection of myocardial fibrosis.57

Furthermore,

ECV represent a newer, more accurate technique of myocardial fibrosis quantification

that is intended to reduce the effects of these variables by using T1 maps of the

myocardium and the blood at three time points (native, 12-and 25-min post-contrast T1

mapping) after correction for hematocrit.26

Based on our findings, we believe there are fundamental differences in the

myocardial response to higher serum FGF-23 concentrations between men and women

consistent with the differences in cardiac remodeling present in response to aging,2

pressure overload,55

volume overload,56

and myocardial infarction.58

Limitations:

ECV values were only measured for a subset of participants (n=588). Only those with

available hematocrit data at the time of CMR examination were used to calculate ECV

(Figure 1). Since only Participants with a GFR ≥45 ml/min (60 ml/min for the site at

Northwestern University) underwent ECV calculation, this subset may represent a

healthier group. Also, since different inclusion criteria were used at different sites as

39

mentioned above, this may introduce heterogeneity. To overcome this we used

multivariate regression to adjust for study site and other potential confounders. We

performed post-hoc stratification by gender. Finally, CMR with T1 mapping was

performed only once at the 5th

follow up exam 10 years after FGF-23 measurement.

Because of this time gap we recognize the potential effect of attrition which may bias our

results. For example, participants with higher FGF-23 are usually sicker and might be

less likely to follow up.

Conclusion:

In a community-based, multi-ethnic cohort free of clinical cardiovascular disease at

baseline, we observed significant associations between higher serum FGF-23

concentrations and higher native myocardial T1 mapping and ECV in men. Our findings

suggest that FGF-23 effects on the myocardium varies by gender and are likely similar to

the effects observed with cardiac remodeling noticed in aging. This mechanism may

explain the accelerated myocardial aging and the higher prevalence of heart failure

observed in patients with CKD. Further studies of FGF-23 effects on the myocardium

may further explore these gender differences.

40

Figure 1. Participant enrollment in the study

Year 2000

N=6814 enrolled in baseline exam

N=6547 had FGF-23 measured at baseline exam

Year 2010-2012

N=1183 with baseline FGF-23 and T1 mapping

N=113 excluded due to visible scar by late

gadolinium enhancement

N=1334 had T1 mapping

N=588 with baseline FGF-23 and ECV

41

Table 1. Baseline characteristics by Fibroblast Growth Factor-23 Quartile

Values are mean ±SD. SBP: Systolic blood pressure. Smoking: current and former versus never. HTN: Hypertension. GFR:

glomerular filtration rate. LVEDM: left ventricular end-diastolic mass (g).

FGF-23 Quartiles

Characteristic overall <31 31-38.2 38.2-46.1 46.1-169.9 P value

No. of subjects 1183 296 296 296 295

Age, y 58±8.7 58±8 57±9 58±9 59±9 0.2

gender 0.004

Female 603 (51%) 173 (58%) 135 (46%) 137 (46%) 158 (54%)

Male 580 (49%) 123 (42%) 161 (54%) 159 (54%) 137 (46%)

Race 0.3

White% 620 (52%) 137 (46%) 154 (52%) 158 (53%) 171 (58%)

Chinese% 140 (12%) 41 (14%) 38 (13%) 31 (11%) 30 (11%)

Black% 260 (22%) 75 (26%) 61 (21%) 64 (22%) 60 (21%)

Hispanic% 163 (14%) 43 (14%) 43 (14%) 43 (14%) 34 (12%)

Diabetes mellitus 176 (15%) 39 (13%) 37 (13%) 46 (16%) 54 (18%) 0.18

Smoking 606 (51%) 161 (54%) 158 (53%) 149 (50%) 138 (47%) 0.2

Education 0.5

High school or less% 304 (26%) 78 (27%) 75 (26%) 86 (29%) 65 (22%)

Some college% 200 (17%) 51 (17%) 45 (15%) 51 (16%) 53 (18%)

College degree or

more%

677 (57%) 165 (56%) 176 (59%) 159 (55%) 177 (60%)

Height (cm) 168±10 167±9 169±9 169±10 169±10 0.1

Weight (kg) 175±37 168±36 177±36 176±36 180±36 <0.001

Body mass index (kg/m2) 28±5 27±5 28±5 28±5 29±5 <0.001

SBP, mm Hg 122±19 122±21 120±19 122±19 123±19 0.3

HTN medication 303 (26%) 76 (26%) 67 (23%) 70 (24%) 90 (31%) 0.13

Estimated GFR,

mL/min per 1.73 m2

84±15 87±17 84±14 84±14 80±15 <0.001

LVEDM (g) 122±32 117±32 125±32 124±33 124±31 0.02

42

Table 2. CMR indices of myocardial fibrosis by gender

Variable Overall Male Female P value

NativeT1 mapping

(ms) 977±43 968±38 985±45 <0.0001

12 min post-contrast

T1 mapping (ms) 457±40 472±33 442±42 <0.0001

25 min post-contrast

T1 mapping (ms) 520±41 535±34 506±41 <0.0001

ECV% 26.9±3.0 25.8±2.8 28±2.8 <0.0001

Values are mean ±SD. ECV: extracellular volume. P value from student t test comparing CMR indices between males and females.

1183 participants had pre- and post-contrast T1 mapping, while 588 participants had ECV measured.

Table 3. Association of FGF-23 with native myocardial T1 mapping, post-contrast T1 mapping,

and extracellular volume

CMR

indices of

myocardial

fibrosis

Gender

FGF-23 Quartiles Linear Model

<31 31-38.2 38.2-46.1 >46.1

Per

doubling of

FGF-23

P value

Native

myocardial

T1 map (ms)

Overall 971±3 974±3 980±3 975±3 2.6 0.3

Men 961±4 966±3 976±3 973±4 7.1 0.053

Women 978±4 983±4 987±4 977±4 0.07 0.9

12-min post-

contrast

myocardial

T1 map (ms)

Overall 458±2 459±2 461±2 457±2 1 0.6

Men 471±3 473±3 473±3 470±3 -1.9 0.5

Women 446±3 443±3 448±3 445±3 2.5 0.4

ECV%

Overall 26.5±0.3 26.9±0.3 26.8±0.3 27±0.2 0.28 0.3

Men 24.7±0.4 25.4±0.3 25.6±0.3 26.3±0.3 1.2 0.007

Women 27.7±0.4 28.2±0.4 28±0.4 27.8±0.4 0.1 0.7

Analysis of variance procedure was used to compare SAS derived least square means of CMR-indices of myocardial fibrosis between

FGF-23 quartiles. For the linear model we used nested multivariate linear regression analysis to examine the association between log2

FGF-23 and CMR indices of myocardial fibrosis. The values are regression coefficient values. For example, for every doubling of

FGF-23 there is an absolute 1.2% increase in ECV. Values are presented as means ± SE. Models were stratified by gender and

adjusted for age, race/ethnicity, study site, education, height and weight; systolic blood pressure, antihypertensive medications, left

ventricular mass, heart rate, low density lipoprotein, high density lipoprotein, diabetes mellitus, smoking, C-reactive protein, urine

albumin-creatinine ratio and eGFRCKD-EPI. P value <0.05 is considered significant. 1183 participants had pre- and post-contrast T1

mapping, while 588 participants had ECV measured

43

Figure 2. Associations between serum FGF-23 quartiles and Native myocardial T1 mapping, 12-

min post-contrast myocardial T1 mapping and ECV, overall (green color) and stratified by

gender, male (blue color) and women (red color).

Overall Men Women

950

955

960

965

970

975

980

985

990

995

Q1 Q2 Q3 Q4

Me

an N

ativ

e T

1 m

aps

(ms)

p-trend=0.02

950

955

960

965

970

975

980

985

990

995

Q1 Q2 Q3 Q4

p-trend=0.002

950

955

960

965

970

975

980

985

990

995

Q1 Q2 Q3 Q4

p-trend=0.13

440

445

450

455

460

465

470

475

480

Q1 Q2 Q3 Q4

Me

an 1

2-m

in p

ost

-co

ntT

1 m

aps

(ms)

p-trend =0.5

440

445

450

455

460

465

470

475

480

Q1 Q2 Q3 Q4

p-trend=0.7

440

445

450

455

460

465

470

475

480

Q1 Q2 Q3 Q4

p-trend=0.8

23

24

25

26

27

28

29

Q1 Q2 Q3 Q4

Me

an E

CV

%

p-trend=0.5

23

24

25

26

27

28

29

Q1 Q2 Q3 Q4

p-trend=0.01

23

24

25

26

27

28

29

Q1 Q2 Q3 Q4

p-trend=0.7

44

SUPPLEMENTAL MATERIAL

Supplemental Table 1: Baseline participant’s characteristics by Fibroblast Growth Factor-23

Quartile for the participants with measured ECV

Values are mean ±SD. SBP: Systolic blood pressure. Smoking: current and former versus never. HTN: Hypertension. GFR:

glomerular filtration rate.

FGF-23 Quartiles

Characteristic Overall <31.8 31.8-39.1 39.1-48.2 48.2-169.9 P value

No. of subjects 588 147 147 147 147

Age, y 58±8.7 57.7±8.2 57.1±8.6 58±9 58.7±9 0.2

Gender 0.3

Female 310 (53%) 87 (59%) 75 (51%) 73 (50%) 75 (51%)

Male 278 (47%) 60 (41%) 72 (49%) 74 (50%) 72 (49%)

Race 0.5

White% 367 (62%) 85 (58%) (93) 63% (94) 64% (95) 65%

Black% 142 (24%) 41 (28%) 35 (24%) 29 (20%) 37 (25%)

Hispanic% 79 (13%) 21 (14%) 19 (13%) 24 (16%) 15 (10%)

Diabetes mellitus 93 (16%) 19 (13%) 17 (12%) 29 (20%) 28 (19%) 0.12

Smoking 320 (54%) 83 (56%) 78 (53%) 80 (54%) 79 (54%) 0.9

Education 0.7

High school or less% 151 (26%) 33 (23%) 44 (30%) 39 (26%) 35 (24%)

Some college% 110 (19%) 28 (19%) 22 (15%) 29 (20%) 31 (21%)

College degree or

more%

325 (55%) 84 (58%) 81 (55%) 79 (54%) 81 (55%)

Height (cm) 169±10 168±9 169±9 169±10 170±11 0.6

Weight (kg) 180±36 175±37 182±35 181±36 184±37 0.16

Body mass index

(kg/m2)

29±5 28±5 29±5 29±5 29±6 0.2

SBP, mm Hg 122±19 123±21 122±18 123±19 123±17 0.9

HTN medication 170 (29%) 43 (29%) 38 (26%) 39 (27%) 50 (34%) 0.4

Estimated GFR,

mL/min per 1.73 m2

82±15 86±16 82±15 80±14 79±15 <0.001

45

1. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation 2000;101:2981-8. 2. Liu CY, Liu YC, Wu C, et al. Evaluation of age-related interstitial myocardial fibrosis with cardiac magnetic resonance contrast-enhanced T1 mapping: MESA (Multi-Ethnic Study of Atherosclerosis). Journal of the American College of Cardiology 2013;62:1280-7. 3. Schwartzkopff B, Frenzel H, Dieckerhoff J, et al. Morphometric investigation of human myocardium in arterial hypertension and valvular aortic stenosis. European heart journal 1992;13 Suppl D:17-23. 4. St John Sutton MG, Lie JT, Anderson KR, O'Brien PC, Frye RL. Histopathological specificity of hypertrophic obstructive cardiomyopathy. Myocardial fibre disarray and myocardial fibrosis. British heart journal 1980;44:433-43. 5. Tyralla K, Amann K. Morphology of the heart and arteries in renal failure. Kidney international Supplement 2003:S80-3. 6. Wong TC, Piehler KM, Kang IA, et al. Myocardial extracellular volume fraction quantified by cardiovascular magnetic resonance is increased in diabetes and associated with mortality and incident heart failure admission. European heart journal 2014;35:657-64. 7. Wong TC, Piehler K, Meier CG, et al. Association between extracellular matrix expansion quantified by cardiovascular magnetic resonance and short-term mortality. Circulation 2012;126:1206-16. 8. Gulati A, Jabbour A, Ismail TF, et al. Association of fibrosis with mortality and sudden cardiac death in patients with nonischemic dilated cardiomyopathy. Jama 2013;309:896-908. 9. Kwong RY, Korlakunta H. Diagnostic and prognostic value of cardiac magnetic resonance imaging in assessing myocardial viability. Topics in magnetic resonance imaging : TMRI 2008;19:15-24. 10. Conrad CH, Brooks WW, Hayes JA, Sen S, Robinson KG, Bing OH. Myocardial fibrosis and stiffness with hypertrophy and heart failure in the spontaneously hypertensive rat. Circulation 1995;91:161-70. 11. Faul C, Amaral AP, Oskouei B, et al. FGF23 induces left ventricular hypertrophy. The Journal of clinical investigation 2011;121:4393-408. 12. Xie J, Cha SK, An SW, Kuro OM, Birnbaumer L, Huang CL. Cardioprotection by Klotho through downregulation of TRPC6 channels in the mouse heart. Nature communications 2012;3:1238. 13. Kestenbaum B, Sachs MC, Hoofnagle AN, et al. Fibroblast growth factor-23 and cardiovascular disease in the general population: the Multi-Ethnic Study of Atherosclerosis. Circulation Heart failure 2014;7:409-17. 14. Lutsey PL, Alonso A, Selvin E, et al. Fibroblast growth factor-23 and incident coronary heart disease, heart failure, and cardiovascular mortality: the Atherosclerosis Risk in Communities study. Journal of the American Heart Association 2014;3:e000936. 15. Saeed M, Wagner S, Wendland MF, Derugin N, Finkbeiner WE, Higgins CB. Occlusive and reperfused myocardial infarcts: differentiation with Mn-DPDP--enhanced MR imaging. Radiology 1989;172:59-64. 16. Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. Journal of the American College of Cardiology 2011;57:891-903. 17. White SK, Sado DM, Flett AS, Moon JC. Characterising the myocardial interstitial space: the clinical relevance of non-invasive imaging. Heart 2012;98:773-9. 18. Bull S, White SK, Piechnik SK, et al. Human non-contrast T1 values and correlation with histology in diffuse fibrosis. Heart 2013;99:932-7.

46

19. Jellis CL, Kwon DH. Myocardial T1 mapping: modalities and clinical applications. Cardiovascular diagnosis and therapy 2014;4:126-37. 20. Jerosch-Herold M, Sheridan DC, Kushner JD, et al. Cardiac magnetic resonance imaging of myocardial contrast uptake and blood flow in patients affected with idiopathic or familial dilated cardiomyopathy. Am J Physiol Heart Circ Physiol 2008;295:H1234-H42. 21. Broberg CS, Chugh SS, Conklin C, Sahn DJ, Jerosch-Herold M. Quantification of diffuse myocardial fibrosis and its association with myocardial dysfunction in congenital heart disease. Circulation Cardiovascular imaging 2010;3:727-34. 22. Neilan TG, Coelho-Filho OR, Shah RV, et al. Myocardial extracellular volume fraction from T1 measurements in healthy volunteers and mice: relationship to aging and cardiac dimensions. JACC Cardiovascular imaging 2013;6:672-83. 23. Bild DE, Bluemke DA, Burke GL, et al. Multi-ethnic study of atherosclerosis: objectives and design. Am J Epidemiol 2002;156:871-81. 24. Iles L, Pfluger H, Phrommintikul A, et al. Evaluation of diffuse myocardial fibrosis in heart failure with cardiac magnetic resonance contrast-enhanced T1 mapping. Journal of the American College of Cardiology 2008;52:1574-80. 25. Schelbert EB, Testa SM, Meier CG, et al. Myocardial extravascular extracellular volume fraction measurement by gadolinium cardiovascular magnetic resonance in humans: slow infusion versus bolus. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2011;13:16. 26. Ugander M, Oki AJ, Hsu LY, et al. Extracellular volume imaging by magnetic resonance imaging provides insights into overt and sub-clinical myocardial pathology. European heart journal 2012;33:1268-78. 27. Messroghli DR, Radjenovic A, Kozerke S, Higgins DM, Sivananthan MU, Ridgway JP. Modified Look-Locker inversion recovery (MOLLI) for high-resolution T1 mapping of the heart. Magnetic resonance in medicine : official journal of the Society of Magnetic Resonance in Medicine / Society of Magnetic Resonance in Medicine 2004;52:141-6. 28. Kellman P, Wilson JR, Xue H, Ugander M, Arai AE. Extracellular volume fraction mapping in the myocardium, part 1: evaluation of an automated method. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2012;14:63. 29. Lewis MR, Callas PW, Jenny NS, Tracy RP. Longitudinal stability of coagulation, fibrinolysis, and inflammation factors in stored plasma samples. Thrombosis and haemostasis 2001;86:1495-500. 30. Imel EA, Peacock M, Pitukcheewanont P, et al. Sensitivity of fibroblast growth factor 23 measurements in tumor-induced osteomalacia. The Journal of clinical endocrinology and metabolism 2006;91:2055-61. 31. Masoudi FA, Havranek EP, Smith G, et al. Gender, age, and heart failure with preserved left ventricular systolic function. Journal of the American College of Cardiology 2003;41:217-23. 32. From AM, Maleszewski JJ, Rihal CS. Current status of endomyocardial biopsy. Mayo Clinic proceedings 2011;86:1095-102. 33. Dall'Armellina E, Piechnik SK, Ferreira VM, et al. Cardiovascular magnetic resonance by non contrast T1-mapping allows assessment of severity of injury in acute myocardial infarction. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2012;14:15. 34. Sibley CT, Noureldin RA, Gai N, et al. T1 Mapping in cardiomyopathy at cardiac MR: comparison with endomyocardial biopsy. Radiology 2012;265:724-32.

47

35. Ellims AH, Iles LM, Ling LH, Hare JL, Kaye DM, Taylor AJ. Diffuse myocardial fibrosis in hypertrophic cardiomyopathy can be identified by cardiovascular magnetic resonance, and is associated with left ventricular diastolic dysfunction. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2012;14:76. 36. Jellis C, Wright J, Kennedy D, et al. Association of imaging markers of myocardial fibrosis with metabolic and functional disturbances in early diabetic cardiomyopathy. Circulation Cardiovascular imaging 2011;4:693-702. 37. Ling LH, Kistler PM, Ellims AH, et al. Diffuse ventricular fibrosis in atrial fibrillation: noninvasive evaluation and relationships with aging and systolic dysfunction. Journal of the American College of Cardiology 2012;60:2402-8. 38. Mongeon FP, Jerosch-Herold M, Coelho-Filho OR, Blankstein R, Falk RH, Kwong RY. Quantification of extracellular matrix expansion by CMR in infiltrative heart disease. JACC Cardiovascular imaging 2012;5:897-907. 39. Rossi MA. Pathologic fibrosis and connective tissue matrix in left ventricular hypertrophy due to chronic arterial hypertension in humans. Journal of hypertension 1998;16:1031-41. 40. Dall'Armellina E, Karia N, Lindsay AC, et al. Dynamic changes of edema and late gadolinium enhancement after acute myocardial infarction and their relationship to functional recovery and salvage index. Circulation Cardiovascular imaging 2011;4:228-36. 41. Brilla CG, Janicki JS, Weber KT. Impaired diastolic function and coronary reserve in genetic hypertension. Role of interstitial fibrosis and medial thickening of intramyocardial coronary arteries. Circulation research 1991;69:107-15. 42. Chan W, Duffy SJ, White DA, et al. Acute left ventricular remodeling following myocardial infarction: coupling of regional healing with remote extracellular matrix expansion. JACC Cardiovascular imaging 2012;5:884-93. 43. Urakawa I, Yamazaki Y, Shimada T, et al. Klotho converts canonical FGF receptor into a specific receptor for FGF23. Nature 2006;444:770-4. 44. Kurosu H, Ogawa Y, Miyoshi M, et al. Regulation of fibroblast growth factor-23 signaling by klotho. The Journal of biological chemistry 2006;281:6120-3. 45. Gutierrez O, Isakova T, Rhee E, et al. Fibroblast growth factor-23 mitigates hyperphosphatemia but accentuates calcitriol deficiency in chronic kidney disease. Journal of the American Society of Nephrology : JASN 2005;16:2205-15. 46. Hu MC, Shi M, Cho HJ, et al. Klotho and phosphate are modulators of pathologic uremic cardiac remodeling. Journal of the American Society of Nephrology : JASN 2015;26:1290-302. 47. Tucker B, Fabbian F, Giles M, Thuraisingham RC, Raine AE, Baker LR. Left ventricular hypertrophy and ambulatory blood pressure monitoring in chronic renal failure. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 1997;12:724-8. 48. Dhingra R, Gaziano JM, Djousse L. Chronic kidney disease and the risk of heart failure in men. Circulation Heart failure 2011;4:138-44. 49. Mall G, Huther W, Schneider J, Lundin P, Ritz E. Diffuse intermyocardiocytic fibrosis in uraemic patients. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 1990;5:39-44. 50. Collins AJ, Foley RN, Chavers B, et al. US Renal Data System 2013 Annual Data Report. American journal of kidney diseases : the official journal of the National Kidney Foundation 2014;63:A7.

48

51. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. The New England journal of medicine 2004;351:1296-305. 52. Edwards NC, Moody WE, Yuan M, et al. Diffuse interstitial fibrosis and myocardial dysfunction in early chronic kidney disease. The American journal of cardiology 2015;115:1311-7. 53. Olivetti G, Giordano G, Corradi D, et al. Gender differences and aging: effects on the human heart. Journal of the American College of Cardiology 1995;26:1068-79. 54. Mallat Z, Fornes P, Costagliola R, et al. Age and gender effects on cardiomyocyte apoptosis in the normal human heart. The journals of gerontology Series A, Biological sciences and medical sciences 2001;56:M719-23. 55. Kostkiewicz M, Tracz W, Olszowska M, Podolec P, Drop D. Left ventricular geometry and function in patients with aortic stenosis: gender differences. International journal of cardiology 1999;71:57-61. 56. Gardner JD, Brower GL, Janicki JS. Gender differences in cardiac remodeling secondary to chronic volume overload. Journal of cardiac failure 2002;8:101-7. 57. Moon JC, Messroghli DR, Kellman P, et al. Myocardial T1 mapping and extracellular volume quantification: a Society for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the European Society of Cardiology consensus statement. Journal of cardiovascular magnetic resonance : official journal of the Society for Cardiovascular Magnetic Resonance 2013;15:92. 58. Biondi-Zoccai GG, Abate A, Bussani R, et al. Reduced post-infarction myocardial apoptosis in women: a clue to their different clinical course? Heart 2005;91:99-101.

49

CHAPTER THREE: ANCILLARY ANALYSES

Fibroblast Growth Factor-23 associations with Incident Heart

Failure with Reduced versus Preserved Ejection Fraction in

the Multi Ethnic Study of Atherosclerosis

Short Title: FGF-23 and HFrEF or HFpEF in MESA.

Mohamed Faher Almahmoud, MD1, Elsayed Z. Soliman MD

1, Alain G. Bertoni, MD

1,

Bryan Kestenbaum, MD2, Ronit Katz, PhD

2, João A. C. Lima, MD

3, Pamela Ouyang,

MD6, P. Elliot Miller, MD

5, Erin D. Michos, MD, M.H.S

4, David M Herrington, MD,

M.H.S1

1Department of Internal Medicine, Section on Cardiology, Wake Forest School of Medicine,

Winston-Salem, NC, USA

2Kidney Research Institute, Department of Medicine, Division of Nephrology, University of

Washington, Seattle, WA, USA

3Department of Radiology, John Hopkins University School of Medicine, Baltimore, MD, USA

4Division of Cardiology, Johns Hopkins University School of Medicine, Baltimore, MD, USA

5Department of Internal Medicine, Division of Hospitalist, John Hopkins University School of

Medicine, Baltimore, MD, USA

6Department of Internal Medicine, Director of women’s cardiovascular health center, John

Hopkins University School of Medicine, Baltimore, MD, USA

Correspondence:

Mohamed Faher Almahmoud MD, Department of Internal Medicine, Section on Cardiology

Wake Forest School of Medicine

Medical Center Boulevard, Winston-Salem, NC

Tel: (336)716.2255, Fax: (336)716.3202

Email: malmahmo@wakehealth.edu

50

Abstract

Background: Fibroblast Growth Factor (FGF)-23 is a hormone produced by osteoblast

as a response to high serum phosphorus. High concentrations of serum FGF-23 are

associated with incident heart failure (HF) in the community. We sought to determine the

association between FGF-23 and incident heart failure with reduced (HFrEF) and with

preserved (HFpEF) ejection fraction.

Methods: We studied 6549 study participants free of heart disease at baseline enrolled in

Multi-Ethnic Study of Atherosclerosis (MESA). FGF-23 concentrations were measured at

baseline between 2000-2002 using Kainos Immunoassay. Individuals were followed for a

mean of 11.1 ± 3 years for incident HF. We classified HF events as HFrEF (ejection

fraction (EF) <50%) or HFpEF (EF ≥50%) at the time of diagnosis. We used Kaplan-

Meier curves and log-rank procedure to compare cumulative incidence of HFrEF and

HFpEF by quartiles of FGF-23. Cox proportional hazard regression was used to compute

hazard ratios (HR) and 95% confidence intervals (CI) for the association between serum

FGF-23 and incident HFrEF and HFpEF. Models were adjusted for age, gender, race,

education, study site, smoking, diabetes, systolic blood pressure, use of antihypertensive

medications, left ventricular (LV) mass, HDL-C, LDL-C, C-reactive protein (CRP),

eGFR, and log urine albumine creatinine ratio.

Results: Study participants had a total of 227 heart failure events. Of these 125 were

classified as HFrEF and 102 as HFpEF. Higher FGF-23 levels were associated with

higher risk for incident HF. For doubling of FGF-23 concentration the adjusted hazard

ratio (HR) was 1.8 (95% CI 1.3-2.5). The association between FGF-23 and HF was

51

stronger with HFpEF compared with HFrEF. Corresponding HRs were 2.6 (95% CI: 2.4-

4) for HFpEF and 1.5 (95% CI; 1-2.3) for HFrEF.

Conclusion: FGF-23 is associated with both HFpEF and HFrEF. We found a stronger

association with HFpEF in the Multi-Ethnic Study of Atherosclerosis. Our findings may

help explain the increase prevalence of HFpEF in patients with chronic kidney disease.

52

Introduction

Heart failure (HF) is a major cause of morbidity and mortality and associated high

health-care related costs.1 It is estimated to affect almost 7 million Americans.

2 Half of

these patients considered to have normal left ventricular ejection fraction (EF), and are

classified as heart failure with preserved ejection fraction (HFpEF).3 Current guidelines

define HFpEF using an EF cut-off of ≥50%.4,5

While the morbidity and mortality of heart

failure with reduced ejection fraction (HFrEF: defined as an EF<50%) is improving,6

HFpEF continues to have unchanged morbidity and mortality.7-9

Although several effective evidence based therapies are now available for the

treatment of patients with HFrEF,4 the same cannot be said about HFpEF.

10 In fact, the

exact pathophysiologic processes that lead to HFpEF are not completely understood yet.11

These processes are diverse and the condition is considered to be multifactorial and

heterogeneous.12,13

Due to this heterogeneity of HFpEF, additional research is needed to

better characterize the phenotype of HFpEF, understand the pathophysiological

processes, as well as identify the risk factors associated with this condition.

Fibroblast growth factor-23 (FGF-23), a hormone involved in phosphorus and

vitamin D hemostasis, 14-16

has been recently linked to the development of HF and left

ventricular hypertrophy (LVH).17,18

In the Mult-Ethnic Study of Atherosclerosis,

Kestenbaum et al. showed that higher serum FGF-23 concentrations are associated with

incident HF, LVH and coronary disease events19

. However, it is unknown whether this

increase in incident HF is of the preserved or reduced ejection fraction types. Even

53

though FGF-23 is strongly linked to LVH which is linked to both types of HF. FGF-23 is

also associated with several other pathologies such as arterial stiffness, endothelial

dysfunction, vascular calcifications as well as major cardiovascular events.20,21

All of

these factors can predispose to both types of HF. We hypothesize that higher levels of

FGF-23 are associated with both HFrEF and HFpEF in a Multi-Ethnic population free of

heart failure at baseline.

Methods

Study population

The Multi-Ethnic Study of Atherosclerosis (MESA) is a prospective cohort study

of cardiovascular disease among 6814 community-living individuals. 22

Between 2000

and 2002, the MESA recruited participants free of cardiovascular disease aged 45 to 84

years from six sites from United States, Baltimore, MD; Chicago, IL; St. Paul, MN;

Forsyth County, NC; New york, NY; and Los Angeles, CA. The MESA recruited a final

population that was 38% white, 28% black, 22% Hispanic, and 12% Asian. The study

excluded any participants with self-reported history of myocardial infarction, angina,

stroke, transient ischemic attack, heart failure, atrial fibrillation, nitroglycerin use,

angioplasty, coronary artery bypass grafting, valve replacement, pacemaker or

defibrillator, or any surgery on the heart or arteries. All participants gave informed

consent, and institutional review board approval was obtained for each site.

We evaluated all participants who had FGF-23 measurement at baseline with

known ejection fraction at the time of developing incident heart failure (n=6549).

54

Study procedures: CMR imaging.

MESA participants without contraindications underwent CMR examinations by

using 1.5-T scanners (Avanto and Espree, Siemens Medical Systems, Eelangen,

Germany) with a 6-channel anterior phased array torso coil and corresponding posterior

coil elements. LV function, dimensions, and myocardial mass were assessed by a cine

steady-state free precession sequence. Twelve short axis slices, one 4-chamber view, and

one 2-chamber view were acquired.

Measurement of Serum FGF-23 Concentrations

Blood and urine samples stored at the University of Vermont Laboratory for

Clinical Biochemistry were shipped on dry ice to the University of Washington where

serum FGF-23 concentrations were measured using the Kainos Immunoassay24

which

detects the full-length biologically intact FGF-23 molecule via mid-molecule and distal

epitopes. Standarized high- and low-value FGF-23 controls within each run were used to

monitor quality control.

Ascertainment of heart failure events

MESA personnel screened participants for incident events through telephone

contacts and scheduled follow up examinations. They abstracted any hospital records

suggesting cardiovascular events. They recorded symptoms, history and biomarkers;

scanned ECGs, echocardiograms, catheterization reports, outpatient records, and other

relevant diagnostic reports; and transmitted these to the coordinating center.

Two study physicians blinded to other study data independently reviewed the

medical records. Incident heart failure events were considered as probable or definite.

55

Definite or probable CHF required heart failure symptoms, such as shortness of breath or

edema, as asymptomatic disease is not a MESA endpoint. In addition to symptoms,

probable CHF required CHF diagnosed by a physician and patient receiving medical

treatment for CHF.

Definite CHF required one or more other criteria, such as pulmonary

edema/congestion by chest X-ray; dilated ventricle or poor LV function by

echocardiography or ventriculography; or evidence of left ventricular diastolic

dysfunction. We considered participants not meeting any criteria, including just a

physician diagnosis of CHF without any other evidence, as having no CHF. The Events

Committee classified heart failure as HFrEF (ejection fraction (EF) <50%) or HFpEF

(EF≥50%) at the time of heart failure diagnosis.

Statistical Analysis

We examined the distribution of the variables and presented continuous variables

as mean ± SD and categorical variables as frequencies and percentages. Basic

characteristics were presented by FGF-23 quartiles. Due to skewed distributions FGF23

was explored as a continuous predictor variable after log base 2 transformations so the

parameter coefficient can be interpreted as “per doubling of FGF23”. We used Kaplan-

Meier curves and the log-rank procedure to compare cumulative incidence of HFrEF and

HFpEF by quartiles of FGF-23. Cox proportional hazard regression was used to compute

cause-specific hazard ratios (HR) and 95% confidence intervals (CI) for the association

between log base 2 transformed FGF-23 and incident HFrEF and HFpEF first, then

between quartiles of FGF-23 and incident HFrEF and HFpEF separately. The initial

56

model was unadjusted. Model 2 was adjusted for age, race, highest education level

completed (high school or less, some college but no degree, college degree or more),

study site and body mass index (BMI). Model 3 was further adjusted for systolic blood

pressure (continuous), antihypertensive medications (yes versus no), LV mass, heart rate,

Low Density Lipoprotein (LDL), High Density Lipoprotein (HDL), diabetes mellitus

status (yes versus no), C-reactive protein (CRP), smoking (Current versus former and

never), urine albumin-creatinine ratio, and eGFRCKD-Epi. P<0.05 was considered

significant for all analyses including interaction terms. All statistical analyses were

performed with SAS software, version 9.3 (SAS Institute, Cary, NC). We tested for the

proportional hazards assumption that the effect of an explanatory variable on the hazard

is constant in time using SAS and we found no evidence of departure from this

assumption.

Results:

Among the 6549 participants (mean age 62±5.5 years), there were 53% women,

39% white, 12% Chinese, 27% black and 22% hispanics. The mean FGF-23 was 40±15

pg/ml. The mean eGFR was 81±18 mL/min/1.73m2.

Compared to participants in the lowest FGF-23 quartile (<31 pg/ml), those in the

highest quartile were older, had higher BMI, systolic blood pressure, lower eGFR, and

more likely to be diabetic (Table 1).

During a mean follow up of 11.1±3 years and a median follow up of 12.1(IQR:

11.6-12.7) years, 227 participants had incident heart failure events. Among these, 125

were classified as HFrEF and 102 were classified as HFpEF. When divided by FGF-23

57

quartiles, Q1 participants had 16 HFrEF and 16 HFpEF events, Q2 participants had 30

HFrEF and 18 HFpEF events, Q3 participants had 37 HFrEF and 24 HFpEF events, and

Q4 had 42 HFrEF and 44 HFpEF events.

We found that higher FGF-23 levels are significantly predictive of future HF

events. In adjusted models, there was an estimated 26% higher risk of HF for every 20

pg/mL increase in FGF-23 (HR 1.26; 95% CI, 1.1-1.5, p value=0.003). Each doubling of

FGF-23 levels was associated with 80% higher risk of incident HF (HR 1.8; 95% CI: 1.3-

2.5, p value <0.001) (Table 2A). We found stronger associations between higher FGF-23

levels and HFpEF events (HR 2.4; 95% CI: 1.4-4, p value <0.001) compared to HFrEF

events (HR 1.5; 95 CI 1-2.3, p value =0.06) for each doubling of FGF-23.

When comparing quartiles of FGF-23, there was a graded significant increase of

the risk of incident HF, incident HFrEF and incident HFpEF. Participants in the highest

FGF-23 quartile had higher risk of HF (HR 2.6; 95%CI; 1.5-4.7), HFrEF (HR 2.3; 95%

CI 1-5), and HFpEF (HR 2.9; 95% CI: 1.2-7.1) compared to participants in the lowest

quartile (reference) (Table 2B, Figure 1). We presented Kaplan-Meier curves with 95%

Hall-Wellner Bands and log-rank test for each outcome in Figure 2, Figure 3A and Figure

3B.

Compared to women, men had -statistically but not clinically significant- higher

FGF-23 concentration (mean FGF-23=40.6 in men compared to 39.6 in women). Because

of the observed difference we repeated the analysis with gender specific FGF-23 quartiles

with minimal cutpoints changes. However, the results were similar.

58

As we found differences between men and women in myocardial response to

higher levels of FGF-23 (in the previous chapter), we repeated the analysis after

stratification by gender. Men had 134 incident HF events with an estimated 60% higher

risk with doubling FGF-23 levels (HR 1.6; 95%CI: 1, 2.5, p value=0.05). In men, we

found significant associations between FGF-23 and each HF types in unadjusted models;

however these associations became non-significant in fully adjusted models. Women had

93 total heart failure events with stronger associations between FGF-23 levels and

incident HF (HR 2.7; 95% CI: 1.5-4.7, p value <0.001) compared to men(supplemental

Table 1). This association was especially stronger for HFpEF (HR 5.2; 95% CI: 2.4-11.7,

p value<0.001).

In sensitivity analyses we performed the analyses without adjusting for LV mass

(Supplemental Table 2A), and again after adjusting for LVH detected by

electrocardiogram (LVH-ECG) instead of LV mass (Supplemental Table 2B) and found

similar results. In additional sensitivity analyses, we adjusted for 24-hour urine phospate

excretion, serum calcium level, serum 1-25 hydroxy vitamin D level, and serum

Parathyroid hormone level and had similar findings.

Discussion:

In a middle-aged multi-ethnic population free of cardiovascular diseae at baseline

(MESA), we demonstrated that higher levels of FGF-23 were associated with higher

incidence of both HFrEF and HFpEF. We also confirmed a previous study done with the

same population but with smaller number of events (n=183) due to less follow up time,

which demonstrated an estimated 19% greater risk of incident HF with each 20 pg/mL

59

increase in FGF-23 (HR 1.19; 95% CI, 1.03-1.37).19

We found similar results. In our

linear model there was an estimated 80% higher risk of incident heart failure with

doubling of FGF-23 levels (Table 2A), and an estimated 26% higher risk for every 20

pg/mL increase in FGF-23 (HR 1.26;95% CI, 1.1-1.5, p value=0.003).

When comparing quartiles we found a graded increase in the risk of all types of

HF. Participants in the highest quartile had at least double the risk of developing HFrEF

and almost 3 times the risk of HFpEF.

We believe the findings of a strong association between FGF-23 with HF can be

explained by the previous observations that found a strong association between FGF-23

with higher LV mass.18,19

However, the reasons why we found stronger associations

between FGF-23 and HFpEF are uncertain. Several previous reports demonstrated that

higher LV mass is associated with greater risk of HF events.24-26

In a study by Velagaleti

et al, eccentric LVH was associated with higher risk of HFrEF, while concentric LVH

was associated with higher risk of HFpEF.24

Furthermore, Seliger et al, demonstrated that

LVH was associated with higher risk of both types of HF especially HFrEF. Although we

adjusted for LV mass in our fully adjusted model, other pathological mechanisms such as

myocardial fibrosis (as we demonstrated in the previous chapter) should be recognize as

another potential step that lead to the development of both HF types.

When comparing risk of heart failure by gender, we found high association

between FGF-23 levels and risk of incident HF in both men and women. The risk for

incident HF was stronger in women than in men. In women, the higher risk of HF was

especially because of strong association with HFpEF. However, our results were limited

60

with wide confidence intervals in these strata because of small number of events. Further

studies are needed to make more accurate conclusions.

To our knowledge this is the first study comparing the associations of FGF-23

with HFrEF and HFpEF. Previous studies examined the association between FGF-23

with ejection fraction as a continuous variable. For example, Kestenbaum et al, found no

significant relationship between FGF-23 and ejection fraction.19

In previous community

based studies, FGF-23 was associated with reduced ejection fraction.21,27

However, these

studies were in participants with chronic kidney disease,21

and in a population undergoing

elective coronary angiography.27

Both of these populations have sicker patients with

more cardiovascular risk factors.

Based on our findings we believe that higher levels of baseline FGF-23 are more

predictive of future HFpEF events. This is important because a significant proportion of

patients with chronic kidney disease suffer from HF especially HFpEF. For example,

50% of ESRD patients have heart failure and up to 85% have abnormal left ventricular

structure and function.28

Since patients with CKD have increasing FGF-23 levels even at

early stages,29,30

it is important to estimate the risk and understand the underlying

mechanism in order to develop new preventive and therapeutic techniques.

Furthermore, our findings of gender differences in the risk of incidence HF with

higher levels of FGF-23 confirm our previous observations of different gender-related

myocardial reaction to higher FGF-23 levels.

61

Limitations:

We only included participants with known EF at the time of HF diagnosis, thus it is

unknown whether the relationship would change had we known the EF of these

participants. Also we are limited by a small number of events to compare both types of

HF so we used definite and probable events which could lead to misclassification error.

However, even with close number of events the association was stronger for HFpEF. We

adjusted for left ventricular mass in our fully adjusted models. This adjustment led to

significant reduction in sample size and loss of power with wider confidence intervals.

However, we found similar results when adjusting for LVH detected by ECG (available

for majority of participants). Nevertheless, left ventricular hypertrophy could be a

mediator and a pathologic mechanism by which higher levels of FGF-23 lead to heart

failure.18

Finally, because of the complexity of FGF-23 physiology, residual confounding

cannot be excluded. For example, we did not adjust for 24-hour urine phosphate

excretion, serum calcium level, vitamin D level, or parathyroid hormone. However, in

sensitivity analyses, adjusting for these covariates didn’t change our results.

Conclusion:

In the Multi-Ethnic Study of Atherosclerosis, higher levels of FGF-23 were associated

with increased risk of incident HFrEF and HFpEF. We observed stronger association

with HFpEF. When comparing the risk of HF by gender, women had stronger risk for

HF, especially HFpEF.

62

Table 1. Baseline Characteristics by Fibroblast Growth Factor-23 1quartile

Values are mean ±SD. SBP: Systolic blood pressure. Smoking: current and former versus never. HTN: Hypertension. GFR: glomerular filtration rate. LV mass: left ventricular mass.

FGF-23 Quartiles

Characteristic overall <31 31-38 38-46 >46 P value

No. of subjects 6549 1639 1638 1635 1637

Age, y 62±10 61±10 62±10 62±10 64±10 <0.001

Gender <0.001

Female 3487(53%) 959(58%) 851(52%) 851(52%) 824(50%)

Male 3062(47%) 680(42%) 784(48%) 784(48%) 813(50%)

Race <0.001

White% 2544(39%) 527(32%) 620(38%) 656(40%) 741(45%)

Chinese% 794(12%) 191(12%) 214(13%) 194(12%) 195(12%)

Black% 1779(27%) 486(30%) 448(27%) 430(26%) 415(25%)

Hispanic% 1432(22%) 435(26%) 356(22%) 355(22%) 286(18%)

Diabetes mellitus 811 (12%) 203(12%) 202(12%) 175(11%) 231(14%) 0.03

Current smoking 843(13%) 274(17%) 218 (13%) 200(12%) 151(9%) <0.001

Education 0.004

High school or less% 2348(36%) 632(39%) 597(36%) 594(37%) 525(32%)

Some college% 1063(16%) 268(16%) 245(15%) 260(16%) 290(18%)

College degree or

more%

3116(48%) 732(45%) 792(49%) 774(47%) 818(50%)

Body mass index

(kg/m2)

28±5.5 27.8±5.4 27.9±5.2 28.5±5.4 29±5.6 <0.001

SBP, mm Hg 126±22 125±22 125±21 126±21 129±22 <0.001

HTN medication 2403(37%) 511(31%) 537(33%) 582(36%) 773(47%) <0.001

Estimated GFR,

mL/min per 1.73 m2

81±18 87±17 83±20 81±17 75±18 <0.001

LV mass (g) 120±29 116±28 120±29 121±30 124±31 <0.001

63

Table 2A. Associations of FGF-23 and incident heart failure, HFrEF and HFpEF

Incident

heart

failure

events

Hazard ratio

(95% CI)

P

value

Incident

HFrEF

events

Hazard ratio

(95% CI)

P

value

Incident

HFpEF

events

Hazard ratio

(95% CI)

P value

Model 1 227 2.1

(1.6-2.7)

<0.001 125 1.8

(1.3-2.5)

0.001 102 2.6

(1.8-3.7)

<0.001

Model 2 226 1.7

(1.3-2.2)

<0.001 124 1.5

(1-2)

0.04 102 2

(1.4-2.9)

<0.001

Model 3 138 1.8

(1.3-2.5)

<0.001 84 1.5

(1-2.3)

0.06 54 2.4

(1.4-4)

0.001

Cox proportional models were used to calculate the hazards ratios of each unit increase in log2 FGF-23 for the development of incident

heart failure, HFrEF and HFpEF. Log2 FGF-23 is interpreted as doubling of FGF-23. For example, for each doubling of FGF-23 there is an increase of 2.4 of the risk of HFpEF. Model 1; unadjusted, Model 2; adjusted for age, gender, race/ethnicity, education, study

site and BMI, Model 3 adjusted for model 2 and systolic blood pressure, antihypertensive medications, LV mass, heart rate, low

density lipoprotein, high density lipoprotein, diabetes mellitus, smoking, C-reactive protein, urine albumin-creatinine ratio and eGFRCKD-EPI. P value <0.05 is considered significant.

Table 2B. Association of FGF-23 and incident heart failure, HFrEF and HFpEF

(Hazards ratios per FGF-23 quartiles)

Fibroblast Growth Factor-23 Quartiles

Q1 Q2 Q3 Q4

Incident heart failure

Model 1 1.0(ref) 1.5(0.96-2.4) 1.9(1.2-2.9) 2.7(1.8-4.1)

Model 2 1.0(ref) 1.3(0.9-2.1) 1.5(1-2.4) 1.9(1.3-2.9)

Model 3 1.0(ref) 1.8(1-3.3) 2.1(1.2-3.8) 2.6(1.5-4.7)

Incident HFrEF

Model 1 1.0(ref) 1.9(1-3.4) 2.2(1.2-4) 2.7(1.5-4.8)

Model 2 1.0(ref) 1.7(0.9-3.1) 1.9(1-3.4) 2(1.2-3.6)

Model 3 1.0(ref) 2(0.9-4.4) 2.8(1.3-5.9) 2.3(1-5)

Incident HFpEF

Model 1 1.0(ref) 1.1(0.6-2.2) 1.5(0.8-2.8) 2.8(1.6-5)

Model 2 1.0(ref) 1(0.5-2) 1.1(0.6-2.2) 1.8(1-3.3)

Model 3 1.0 (ref) 1.6(0.6-4.1) 1.3 (0.5-3.3) 2.9 (1.2-7.1)

Cox proportional model was used to calculate the hazards ratios. Values presented are hazards ratios and (95% confidence intervals)

for each quartile of FGF-23 compared to reference group(Q1). Model 1; unadjusted, Model 2; adjusted for age, gender, race/ethnicity, education, study site and BMI, Model 3 adjusted for model 2 and systolic blood pressure, antihypertensive medications, LV mass,

heart rate, low density lipoprotein, high density lipoprotein, diabetes mellitus, smoking, C-reactive protein, urine albumin-creatinine

ratio and eGFRCKD-EPI. P value <0.05 is considered significant.

64

Figure 1. Hazard ratios of the associations of FGF-23 quartiles with incident heart failure, HFrEF

and HFpEF

Figure 2. Higher FGF-23 levels are associated with increased risk of incident heart failure.

Participants were stratified into quartiles of FGF-23. (Kaplan-Meier curve with 95% Hall-Wellner

Brands and log-rank test). Notice the Y axis start from 90% survival, X axis is the time to heart

failure event (days).

0

0.5

1

1.5

2

2.5

3

3.5

q1 q2 q3 q4

Haz

arad

rat

io p

er

FGF-

23

qu

arti

les

Incident Heart Failure

0

0.5

1

1.5

2

2.5

3

3.5

q1 q2 q3 q4

Incident HFrEF

0

0.5

1

1.5

2

2.5

3

3.5

q1 q2 q3 q4

Incident HFpEF

65

Figure 3A. Higher FGF-23 levels are associated with increased risk of incident HFrEF.

Participants were stratified into quartiles of FGF-23. (Kaplan-Meier curve with 95% Hall-Wellner

Bands and log-rank test).

Figure 3B. Higher FGF-23 levels are associated with increased risk of incident HFpEF.

Participants were stratified into quartiles of FGF-23. (Kaplan-Meier curve with 95% Hall-Wellner

Bands and log-rank test).

66

SUPPLEMENTAL MATERIAL

Supplemental Table 1. Associations of FGF-23 and incident heart failure, HFrEF and HFpEF

stratified by gender

Incident

heart

failure

events

Hazard

ratio (95%

CI)

P value Incident

HFrEF

events

Hazard

ratio (95%

CI)

P value Incident

HFpEF

events

Hazard

ratio (95%

CI)

P value

Men

Model 1 134 1.9

(1.4-2.7)

<0.001 86 1.8

(1.1-2.7)

0.01 48 2.3

(1.3-4)

0.004

Model 2 133 1.6

(1.1-2.2)

0.008 85 1.5

(1-2.3)

0.054 48 1.8

(1-3.2)

0.056

Model 3 86 1.6

(1-2.5)

0.05 59 1.6

(0.9-2.7)

0.1 27 1.7

(0.7-3.9)

0.2

Women

Model 1 93 2.3

(1.6-3.2)

<0.001 39 1.7

(0.9-3)

0.09 54 2.8

(1.8-4.4)

<0.001

Model 2 93 1.8

(1.2-2.6)

<0.001 39 1.3

(0.7-2.5)

0.4 54 2.2

(1.3-3.7)

0.003

Model 3 52 2.7

(1.5-4.7)

<0.001 25 1.7

(0.7-4.2)

0.24 27 5.2

(2.4-11.7)

<0.001

Cox proportional models were used to calculate the hazards ratios of each unit increase in log2 FGF-23 for the development of incident

heart failure, HFrEF and HFpEF. Log2 FGF-23 is interpreted as doubling of FGF-23. For example, for each doubling of FGF-23 there

is an increase of 5.2 of the risk of HFpEF in women. Model 1; unadjusted, Model 2; adjusted for age, race/ethnicity, education, study site and BMI, Model 3 adjusted for model 2 and systolic blood pressure, antihypertensive medications, LV mass, heart rate, low

density lipoprotein, high density lipoprotein, diabetes mellitus, smoking, C-reactive protein, urine albumin-creatinine ratio and

eGFRCKD-EPI. P value <0.05 is considered significant.

Supplemental Table 2A. Associations of FGF-23 and incident heart failure, HFrEF and HFpEF

overall and stratified by gender without adjustment for LV mass

Model 3 Incident

heart

failure

events

Hazard

ratio (95%

CI)

P value Incident

HFrEF

events

Hazard

ratio (95%

CI)

P value Incident

HFpEF

events

Hazard

ratio (95%

CI)

P value

Overall 217 1.6

(1.2-2.1)

<0.001 121 1.3

(0.9-1.9)

0.13 96 2.1

(1.4-3.1)

<0.001

Men 129 1.5

(1-2.1)

0.03 82 1.5

(0.8-2.3)

0.11 47 1.5

(0.8-2.8)

0.17

Women 88 1.9

(1.3-3)

0.003 39 1.2

(0.6-2.2)

0.7 49 2.9

(1.6-5.3)

<0.001

Cox proportional models were used to calculate the hazards ratios of each unit increase in log2 FGF-23 for the development of incident

heart failure, HFrEF and HFpEF. Log2 FGF-23 is interpreted as doubling of FGF-23. For example, for each doubling of FGF-23 there

is an increase of 2.9 of the risk of HFpEF in women. Model 3 adjusted for age, race/ethnicity, education, study site, BMI, systolic

blood pressure, antihypertensive medications, heart rate, low density lipoprotein, high density lipoprotein, diabetes mellitus, smoking,

C-reactive protein, urine albumin-creatinine ratio and eGFRCKD-EPI. P value <0.05 is considered significant.

67

Supplemental Table 2B. Associations of FGF-23 and incident heart failure, HFrEF and HFpEF

overall and stratified by gender with adjustment for LVH-ECG

Model 3 Incident

heart

failure

events

Hazard

ratio (95%

CI)

P value Incident

HFrEF

events

Hazard

ratio (95%

CI)

P value Incident

HFpEF

events

Hazard

ratio (95%

CI)

P value

Overall 217 1.6

(1.2-2.1)

<0.001 121 1.3

(0.9-1.8)

0.17 96 2.1

(1.4-3.1)

<0.001

Men 129 1.5

(1-2.1)

0.045 82 1.4

(0.9-2.2)

0.13 47 1.5

(0.8-2.8)

0.19

Women 88 1.9

(1.3-3)

0.003 39 1.1

(0.6-2.2)

0.7 49 2.9

(1.6-5.2)

<0.001

Cox proportional models were used to calculate the hazards ratios of each unit increase in log2 FGF-23 for the development of incident heart failure, HFrEF and HFpEF. Log2 FGF-23 is interpreted as doubling of FGF-23. For example, for each doubling of FGF-23 there

is an increase of 2.9 of the risk of HFpEF in women. Model 3 adjusted for age, race/ethnicity, education, study site, BMI, systolic

blood pressure, antihypertensive medications, heart rate, LVH-ECG (left ventricular hypertrophy detected by electrocardiogram), low

density lipoprotein, high density lipoprotein, diabetes mellitus, smoking, C-reactive protein, urine albumin-creatinine ratio and

eGFRCKD-EPI. P value <0.05 is considered significant.

68

1. Schocken DD, Benjamin EJ, Fonarow GC, et al. Prevention of heart failure: a scientific statement from the American Heart Association Councils on Epidemiology and Prevention, Clinical Cardiology, Cardiovascular Nursing, and High Blood Pressure Research; Quality of Care and Outcomes Research Interdisciplinary Working Group; and Functional Genomics and Translational Biology Interdisciplinary Working Group. Circulation 2008;117:2544-65. 2. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics--2012 update: a report from the American Heart Association. Circulation 2012;125:e2-e220. 3. Steinberg BA, Zhao X, Heidenreich PA, et al. Trends in patients hospitalized with heart failure and preserved left ventricular ejection fraction: prevalence, therapies, and outcomes. Circulation 2012;126:65-75. 4. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. European journal of heart failure 2012;14:803-69. 5. Yancy CW, Jessup M, Bozkurt B, et al. 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Journal of the American College of Cardiology 2013;62:e147-239. 6. Loh JC, Creaser J, Rourke DA, et al. Temporal trends in treatment and outcomes for advanced heart failure with reduced ejection fraction from 1993-2010: findings from a university referral center. Circulation Heart failure 2013;6:411-9. 7. Chan MM, Lam CS. How do patients with heart failure with preserved ejection fraction die? European journal of heart failure 2013;15:604-13. 8. Owan TE, Hodge DO, Herges RM, Jacobsen SJ, Roger VL, Redfield MM. Secular trends in renal dysfunction and outcomes in hospitalized heart failure patients. Journal of cardiac failure 2006;12:257-62. 9. Yancy CW, Lopatin M, Stevenson LW, et al. Clinical presentation, management, and in-hospital outcomes of patients admitted with acute decompensated heart failure with preserved systolic function: a report from the Acute Decompensated Heart Failure National Registry (ADHERE) Database. Journal of the American College of Cardiology 2006;47:76-84. 10. Shah AM, Pfeffer MA. The many faces of heart failure with preserved ejection fraction. Nature reviews Cardiology 2012;9:555-6. 11. Oktay AA, Rich JD, Shah SJ. The emerging epidemic of heart failure with preserved ejection fraction. Curr Heart Fail Rep 2013;10:401-10. 12. Bhuiyan T, Maurer MS. Heart Failure with Preserved Ejection Fraction: Persistent Diagnosis, Therapeutic Enigma. Current cardiovascular risk reports 2011;5:440-9. 13. Cleland JG, Pellicori P. Defining diastolic heart failure and identifying effective therapies. Jama 2013;309:825-6. 14. Larsson T, Marsell R, Schipani E, et al. Transgenic mice expressing fibroblast growth factor 23 under the control of the alpha1(I) collagen promoter exhibit growth retardation, osteomalacia, and disturbed phosphate homeostasis. Endocrinology 2004;145:3087-94. 15. Gattineni J, Baum M. Regulation of phosphate transport by fibroblast growth factor 23 (FGF23): implications for disorders of phosphate metabolism. Pediatric nephrology 2010;25:591-601. 16. Shimada T, Hasegawa H, Yamazaki Y, et al. FGF-23 is a potent regulator of vitamin D metabolism and phosphate homeostasis. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 2004;19:429-35.

69

17. Ix JH, Katz R, Kestenbaum BR, et al. Fibroblast growth factor-23 and death, heart failure, and cardiovascular events in community-living individuals: CHS (Cardiovascular Health Study). Journal of the American College of Cardiology 2012;60:200-7. 18. Faul C, Amaral AP, Oskouei B, et al. FGF23 induces left ventricular hypertrophy. The Journal of clinical investigation 2011;121:4393-408. 19. Kestenbaum B, Sachs MC, Hoofnagle AN, et al. Fibroblast growth factor-23 and cardiovascular disease in the general population: the Multi-Ethnic Study of Atherosclerosis. Circulation Heart failure 2014;7:409-17. 20. Desjardins L, Liabeuf S, Renard C, et al. FGF23 is independently associated with vascular calcification but not bone mineral density in patients at various CKD stages. Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA 2012;23:2017-25. 21. Scialla JJ, Xie H, Rahman M, et al. Fibroblast growth factor-23 and cardiovascular events in CKD. Journal of the American Society of Nephrology : JASN 2014;25:349-60. 22. Bild DE, Bluemke DA, Burke GL, et al. Multi-ethnic study of atherosclerosis: objectives and design. Am J Epidemiol 2002;156:871-81. 23. Imel EA, Peacock M, Pitukcheewanont P, et al. Sensitivity of fibroblast growth factor 23 measurements in tumor-induced osteomalacia. The Journal of clinical endocrinology and metabolism 2006;91:2055-61. 24. Velagaleti RS, Gona P, Pencina MJ, et al. Left ventricular hypertrophy patterns and incidence of heart failure with preserved versus reduced ejection fraction. The American journal of cardiology 2014;113:117-22. 25. Bluemke DA, Kronmal RA, Lima JA, et al. The relationship of left ventricular mass and geometry to incident cardiovascular events: the MESA (Multi-Ethnic Study of Atherosclerosis) study. Journal of the American College of Cardiology 2008;52:2148-55. 26. Seliger SL, de Lemos J, Neeland IJ, et al. Older Adults, "Malignant" Left Ventricular Hypertrophy, and Associated Cardiac-Specific Biomarker Phenotypes to Identify the Differential Risk of New-Onset Reduced Versus Preserved Ejection Fraction Heart Failure: CHS (Cardiovascular Health Study). JACC Heart failure 2015;3:445-55. 27. Seiler S, Cremers B, Rebling NM, et al. The phosphatonin fibroblast growth factor 23 links calcium-phosphate metabolism with left-ventricular dysfunction and atrial fibrillation. European heart journal 2011;32:2688-96. 28. Foley RN, Parfrey PS, Sarnak MJ. Epidemiology of cardiovascular disease in chronic renal disease. Journal of the American Society of Nephrology : JASN 1998;9:S16-23. 29. Gutierrez OM, Mannstadt M, Isakova T, et al. Fibroblast growth factor 23 and mortality among patients undergoing hemodialysis. The New England journal of medicine 2008;359:584-92. 30. Isakova T, Wahl P, Vargas GS, et al. Fibroblast growth factor 23 is elevated before parathyroid hormone and phosphate in chronic kidney disease. Kidney international 2011;79:1370-8.

70

CURRICULUM VITAE

Date Prepared: 04/01/2016

Name: Mohamed Faher Almahmoud

Office Address:

Medical Center Boulevard

Section on Hospital Medicine

Winston Salem, NC 27157-1021

Home Address:

6164 Chamberlain PL, Apt 301

Winston Salem, NC 27103

Work Phone: 336.716.2255

Cell Phone: 201-310-8484

Work Email:

malmahmo@wakehealth.edu

Work FAX: 336.716.3202

Place of Birth: Aleppo, Syria

Education

2010 M.D. Medicine University of Aleppo

2014 Residency, Internal Medicine at St. John Hospital/Wayne State University

Professional Societies

2015

2015

2003

American Heart Association

American College of Cardiology

Syrian students union organization

Member

Member

Member

2010 Syrian Ministry of Physicians Member

2011 American College of Physicians Member

71

Honors and Prizes

2013 Poster Presentation

award

American College of

Physicians

2nd

place poster

presentation at the

associate day conference

of ACP-Michigan chapter

2014 Outstanding Resident Research Award, St. John Hospital Medical Center

Report of Local Teaching and Training

Teaching of Students in Courses

2011-2013 Instructor of Internal Medicine Wayne State University Hospital

2014-2015

2014-2016

Instructor of Internal Medicine

Cardiovascular research

fellowship

Wake Forest Baptist Medical Center

Wake Forest University

Report of Regional, National and International Invited

Teaching and Presentations

Invited Presentations and Courses

Regional

2012 Propionibacterium acne cerebral abscess following craniotomy

Almahmoud, Mohamed Faher., Anilrudh, Venugopal., Leonard, Johnson.

Oral presentation at the American College of Physicians-Michigan

chapter associate meeting

2012 Outcomes of correcting hyponatremia in patients with myocardial

infarction

Almahmoud, Mohamed Faher., Qureshi, Waqas., Alirhayim, Zaid.,

Khalid, Fatima

Research presentation at Michigan state medical society annual scientific

meeting

2013 H1N1-induced encephalitis in an adult patient

Almahmoud, Mohamed Faher., Obeid, Karam., Johnson, Leonard.

Poster presentation at the American College of Physicians-Michigan

associate meeting

2013 Isolated Mycobacterium kansasii Cellulitis and Osteomyelitis in an

Immunocompetent Patient

Almahmoud, Mohamed Faher., Obeid, Karam., Johnson, Leonard.

Poster presentation at the American College of Physicians-Michigan

associate meeting

72

2013 Seizure induced stress cardiomyopathy in an epileptic patient

Almahmoud, Mohamed Faher., Qaqish, Nicole., Michael, Frederick.

Poster presentation at the American College of Physicians-Michigan

chapter annual scientific meeting

2013

2014

2014

2014

2014

Failure Rates After Arteriovenous Fistula Creations

Ali, Sajid., Almahmoud, Mohamed Faher., Bellovic, Keith.

Poster abstract presentation at the American College of Physicians-

Michigan chapter annual scientific meeting

Outcomes of Laser-Assisted Balloon Angioplasty versus Balloon

Angioplasty Alone for Below Knee Peripheral Arterial Disease

Piyaskulkaew, Chatchawan ., Parvataneni, Kesav., Ballout, Hussien.,

Sharma, Tarun., Almahmoud, Mohamed., Ketron, Lowell., Szpunar, Susan

., LaLonde, Thomas, Mehta, Rajendra., Yamasaki, Hiroshi.

Oral presentation at the American College of Physicians-Michigan

associate meeting

Myxedema Coma in a Patient with End Stage Renal Disease on Thyroid

Replacement Therapy

Almahmoud, Mohamed Faher., Alamiri, Khaled., Russel, Sherilyn.

Poster presentation at the American College of Physicians-Michigan

associate meeting

Moyamoya Syndrome in a 28-year old female with Systemic Lupus

Erythematous

Hassanali, Alia., Almahmoud, Mohamed Faher., Qaqish, Nicole..

Poster presentation at the American College of Physicians-Michigan

associate meeting

Implications of Transesophageal Echocardiography on the Diagnosis and

Management of Implantable Cardiac Device Infections

Almahmoud, Mohamed Faher., Fishbain, Joel., Othman, Hussein.,

Sabbagh, Salah., Szpunar, Susan

Poster Presentation at St.John hospital Annual research day

National

2012 Hemophagocytic lymphohistocytosis syndrome

Almahmoud, Mohamed Faher., Andreeff, Michael.

Oral presentation at the weekly meeting of the leukemia

department at MD Anderson Hospital, University of Texas

in Houston, Texas

73

2012 Incidence of Arterial and Venous Thrombosis in Von Willebrand Disease

2014

2015

2015

2015

2016

Hassan, Syed., Qureshi, Waqas., Amer, Syed., Almahmoud, Mohamed.,

Dabak, Vrushali S., Kuriakose, Philip.

Oral presentation at the 2012 Annual Meeting of the American Society of

Hematology in Atlanta, Georgia

Outcomes of Laser-Assisted Balloon Angioplasty versus Balloon

Angioplasty Alone for Below Knee Peripheral Arterial Disease

American College of Cardiology 63rd Annual Scientific Session and

TCT@ACC-i2, Washington, DC, USA

J Am Coll Cardiol. 2014;63(12_S):. doi:10.1016/S0735-1097(14)62142-2

The Role of Transesophageal Echocardiography in Extraction of Infected

Cardiac Implantable Electrophysiological Devices

Almahmoud, Mohamed Faher., Fishbain, Joel., Othman, Hussein.,

Sabbagh, Salah., Szpunar, Susan..

American College of Cardiology 64th

Annual Scientific Session and

TCT@ACC-i2,

San Diego, CA, USA

J Am Coll Cardiol. 2015;65(10_S):. doi:10.1016/S0735-1097(15)60320-5

Electrocardiographic versus Echocardiographic Left Ventricular

Hypertrophy in Prediction of Congestive Heart Failure in the Elderly

Almahmoud, Mohamed Faher., Soliman, Elsayed., Qureshi, Waqas.,

O'neal, Wesley.

Moderated poster abstract presentation

American Heart Association, Epi/lifestyle Scientific Sessions, Baltimore,

MD, USA

Circulation.2015; 131: AMP55

Resting heart rate and incident atrial fibrillation in the elderly

Almahmoud, Mohamed Faher., Soliman, Elsayed., O'neal, Wesley.

Poster abstract presentation

American Heart Association, Epi/lifestyle Scientific Sessions, Baltimore,

MD, USA

Circulation.2015; 131: AP102

Left Ventricular Hypertrophy by Electrocardiogram versus Cardiac

Magnetic Resonance Imaging as a Predictor for Heart Failure: The MESA

study

Oseni, Abdullahi O., Qureshi, Waqas T., Almahmoud, Mohamed Faher.,

Bertoni, Alain., Bluemke, David A., Hundley, William G., Lima, Joao

A.C., Herrington, David M., Soliman, Elsayed Z.

Moderated poster presentation

American Heart Association, Epi/lifestyle Scientific Sessions, Phoenix,

AZ, USA. Circulation.2016;133: AMP31

74

2016 Fibroblast Growth Factor-23 and Cardiac MRI indices of Myocardial

Fibrosis in the Multi-Ethnic Study of Atherosclerosis

Almahmoud, Mohamed Faher., Herrington, David., Hundley, W., Jordan,

Jennifer., Kestenbaum, Bryan., Katz, Ronit., Michos, Erin., Liu, Chia-

Ying., Venkatesh, Bharath Ambale., Lima, Joao.

American College of Cardiology 65th

Annual scientific session & Expo,

2016, Chicago, IL, USA

International

2008 Updates in Rheumatoid Arthritis management

Almahmoud, Mohamed Faher.

Oral presentation at the University of Aleppo 50th Anniversary Meeting.

Report of Clinical Activities and Innovations

Current Licensure and Certification

2010 Educational Commission for Foreign Medical Graduates

2014

Controlled Substance Registration Certificate

2014

2014

2015

2015

2015

North Carolina Medical Board, Physician License Certificate

Council on Epidemiology and Prevention, Participation as a Fellow in the

40th

Ten-Day Seminar on the Epidemiology and Prevention of

Cardiovascular Disease

July 28-August 8, 2014

American Heart Association Basic Life Support (BLS) for healthcare

providers (CPR and AED) program.

May 20th

, 2017

American Heart Association Advanced Cardiovascular Life Support

(ACLS) program.

May 20th

, 2017

West Virginia Medical Board, Physician License Certificate

75

Practice Activities

2014

2014

2014

Cardiovascular Research Fellow at Wake

Forest University, School of Medicine

Master of Science in Clinical and

Population Translational Sciences at

Wake Forest University, School of

Medicine

Instructor of Internal Medicine at Wake

Forest Baptist Hospital

Report of Education of Patients and Service to the Community

Activities

2012

2015

Diabetic Educations at St. John Hospital Diabetes Clinic

Role: educating patients about current management of diabetes

Triad Health free community clinic

Role: volunteering primary physician

Report of Scholarship

Publications

1- Resting heart rate and incident atrial fibrillation in the elderly

O'neal, Wesley T., Almahmoud, Mohamed Faher., Soliman, Elsayed Z..

Pacing Clin Electrophysiol. 2015 May;38(5):591-7. doi: 10.1111/pace.12591. Epub

2015 Feb 17

2- The Role of Transesophageal Echocardiography in Extraction of Infected

Cardiac Implantable Electrophysiological Devices

Almahmoud, Mohamed Faher., Fishbain, Joel., Othman, Hussein., Sabbagh, Salah.,

Szpunar, Susan.

Abstract: J Am Coll Cardiol. 2015;65(10_S):. doi:10.1016/S0735-1097(15)60320-5

3- Outcomes of Laser-Assisted Balloon Angioplasty versus Balloon Angioplasty

Alone for Below Knee Peripheral Arterial Disease

76

Piyaskulkaew,Chatchawan., Parvataneni, Kesav., Ballout, Hussien., Sharma, Tarun.,

Almahmoud, Mohamed.,Ketron, Lowell., Szpunar, Susan., LaLonde, Thomas., Mehta,

Rajendra., Yamasaki, Heroshi.

Abstract: J Am Coll Cardiol. 2014;63(12_S):. doi:10.1016/S0735-1097(14)62142-2

4- Electrocardiographic and Echocardiographic Left Ventricular Hypertrophy

in the Prediction of Stroke in the Elderly

O'neal, Wesley T., Almahmoud, Mohamed Faher., Qureshi, Waqas T., Soliman, Elsayed

Z..

J Stroke Cerebrovasc Dis. 2015 Jul 4. pii: S1052-3057(15)00314-6. doi:

10.1016/j.jstrokecerebrovasdis.2015.04.044

5- Electrocardiographic versus Echocardiographic Left Ventricular

Hypertrophy in Prediction of Congestive Heart Failure in the Elderly

Almahmoud, Mohamed Faher., O'neal, Wesley T., Qureshi, Waqas T., Soliman, Elsayed

Z..

Clin. Cardiol. 38, 000-000 (2015) DOI:10.1002/clc.22402

6- Outcomes of Correcting Hyponatremia in Patients with Myocardial

Infarction

Qureshi, Waqas., Hassan, Syed., Khalid, Fatima., Almahmoud, Mohamed Faher., Shah,

Bhavik., Tashman, Ra'ad., Ambulgekar, Nikhil., El-Refai, Mostafa., Mittal, Chetan.,

Alirhayim, Zaid..

Clin Res Cardiol. 2013 Sep;102(9):637-44. doi: 10.1007/s00392-013-0576-z. Epub 2013

May 8.

7- Incidence of Arterial and Venous Thrombosis in Vonwillebrand Disease

Hassan, Syed. Qureshi, Waqas., Amer, Syed., Almahmoud, Mohamed., Dabak, Vrushali

S., Kuriakose,Philip.

322. Disorder of Coagulation or Fibrinolysis I, Blood (ASH annual Meeting Abstracts)

2012 120: Abstract 101

AMER SOC HEMATOLOGY 2021 L ST NW, SUITE 900, WASHINGTON, DC 20036 USA

Conference: BLOOD, Volume: 21

8- Laser in infrapopliteal and popliteal stenosis 2 study (LIPS2): Long-term

outcomes of laser-assisted balloon angioplasty versus balloon angioplasty for below

knee peripheral arterial disease

Piyaskulkaew, Chatchawan., Parvataneni, Kesav., Ballout Hussein., Szpunar, Susan.,

Sharma, Tarun., Almahmoud, Mohamed., Davis, Thomas., Mehta, Rajendra H.,

Yamasaki, Hiroshi. Catheter Cardiovasc Interv. 2015 Dec 1;86(7):1211-8. doi: 10.1002/ccd.26145. Epub 2015 Oct 22.

9- Left Ventricular Hypertrophy by Electrocardiogram versus Cardiac

Magnetic Resonance Imaging as a Predictor for Heart Failure: The MESA study

77

Oseni, Abdullahi O., Qureshi, Waqas T., Almahmoud, Mohamed Faher., Bertoni, Alain.,

Bluemke, David A., Hundley, William G., Lima, Joao A.C., Herrington, David M.,

Soliman, Elsayed Z.

Moderated poster presentation

American Heart Association, Epi/lifestyle Scientific Sessions, Phoenix, AZ, USA

Circulation.2016;133: AMP31

10- Assisted Reproductive Techniques

Almahmoud, Mohamed Faher., Mohsen, Mousa. (2009, June 10).

Book, Aleppo Publications, Pub Status: Published.

Professional educational materials or reports, in print or other media

1- Heparin Induced Thrombocytopenia in Patients with Cancer

Clinical Research

2- The Role of Transesophageal Echocardiography in Extraction of Infected Cardiac

Implantable Electrophysiological Devices

Clinical Research

3- Failure rates after Arteriovenous Fistula creations

Clinical Research

4- Changes in Plasma Volume as a Prognostic Marker in Myocardial Infarction

Clinical Research

5- Manuscript Reviewer for Frontiers in Medicine Journal 2014:

Cardiovascular epidemiology

6- Fibroblast Growth Factor 23 and Myocardial Fibrosis in Multi-Ethnic Study of

Atherosclerosis using Cardiac Magnetic Resonance T1-Mapping

Accepted proposal

7- Relationship Between Cardiac Structure, Function and Mechanics with

Neurocognition in the Hisopanic Community Health Study/Study of Lations (SOL)

Echocardiographic Study of Lations (Echo-SOL) Ancillary Study

Accepted proposal

8- Electrocardiographic versus Cardiac Magnetic Resonance Left Ventricular

Hypertrophy in Prediction of Congestive Heart Failure in the Mutli-Ethnic Study of

Atherosclerosis

Accepted proposal

78

9- I was an Ad hoc reviewer for Catheterization and Cardiovascular Interventions

Journal. 2015

Work Experience

11/2012 - 12/2012 Average Hours/Week: 60

MD Anderson Hospital/University of Texas, Texas Rotating Resident, with Dr. Michael Andreeff

I worked as a rotating resident at the leukemia department at MD Anderson Hospital.

During this rotation, I had exposure to inpatient and outpatient clinical cases of

leukemia and attended conferences and lectures about emerging researches in the

field of oncology and leukemia at MD Anderson hospital; also I did a presentation

about hemophagocytic lymphohistocytosis syndrome at the leukemia department.

03/2011 - 06/2011

Al Fatih Hospital, Syrian Arab Republic ER Physician, with Dr. Mahmoud, Haj Mohamed

I worked in the ER Department as an assistant physician.

During this period, I managed medical, surgical and pediatric emergencies.

12/2009 - 01/2010

Suny Downstate University, New York Medical Student, with Dr. Nicholas Shorter

I rotated as a medical student in the Pediatric Surgery Department with Dr.Shorter.

I helped in all pediatric operations and interventions at King's County Hospital and

Downstate Hospital. This was an elective that I did during my medical education.

11/2009 - 12/2009

Suny Downstate University, New York Medical Student, with Dr. JEFFREY WEISS

I rotated as a medical student in the department of Urology with Dr.Weiss. I

participated in inpatient and outpatient urological services.

I also helped in many different Urological operations and interventions.

10/2008 - 11/2008

American University of Beirut, Lebanon Medical student, with Dr. Mahmoud Choucair

I rotated as a medical student elective at the department of Internal Medicine at the

American University of Beirut Hospital, where I had exposure to inpatient and

outpatient general medicine, cardiology and neurology patients.

79

Volunteer Experience 12/ 2010 - 01/2011

Wayne state University, Michigan Rotating Physician, with Dr. Theodore Jones

I worked at Hutzel Women's Hospital in the department of OB/GYN as a rotating

physician.

I rotated mainly in the Department of Labor and delivery, and also helped in the

OB/GYN operations.

10/2014- 01/2015

Wake Forest University School of Medicine

Participated in the WFU School of Medicine admission interviews.

Current/Prior Training 07/2015

Cardiovascular research fellowship, NIH T32 training grant (Mentor- David

Herrington), Wake Forest School of Medicine, July 2015-June 2016

07/28-08/08, 2014

Council on Epidemiology and Prevention, Participation as a Fellow in the 40th

Ten-Day Seminar on the Epidemiology and Prevention of Cardiovascular

Disease Lake Tahoe, CA, USA

07/2011 – 06/2014

Internal Medicine residency

St John Hospital and Medical Center Program, Grosse pointe Woods, Michigan Internal Medicine

Raymond C Hilu

Rozzell, Donald

11/2012 - 12/2012

Internal Medicine residency

Md Anderson Cancer center/university of Texas, Houston, Texas

Internal Medicine

Philip R Orlander

Michael Andreeff

Other Awards/Accomplishments

80

1-Syrian Youth organization award and medal for first place in Arabesque art 1994

2-Aleppo Youth organization award and medal for first place in Arabesque and

Arabic calligraphy 1994-1995-1996

3-University of Aleppo soccer championship 2006 second place

4-University of Aleppo soccer championship 2007 first place

5-University of Aleppo soccer championship 2008 second place

6-University of Aleppo soccer championship 2009 first place

7-Medical school soccer championship 2010 First place

8-Medical school soccer championship top scorer 2010

9-Top 10 Medical students honor