Cardiac Fibrosis in Patients With Atrial Fibrillation

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 66, NO. 8, 2015

ª 2015 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER INC.

http://dx.doi.org/10.1016/j.jacc.2015.06.1313

THE PRESENT AND FUTURE STATE-OF-THE-ART REVIEW

Cardiac Fibrosis in Patients With Atrial Fibrillation Mechanisms and Clinical Implications Mikhail S. Dzeshka, MD,*y Gregory Y.H. Lip, MD,*z Viktor Snezhitskiy, PHD,y Eduard Shantsila, PHD*

ABSTRACT Atrial fibrillation (AF) is associated with structural, electrical, and contractile remodeling of the atria. Development and progression of atrial fibrosis is the hallmark of structural remodeling in AF and is considered the substrate for AF perpetuation. In contrast, experimental and clinical data on the effect of ventricular fibrotic processes in the pathogenesis of AF and its complications are controversial. Ventricular fibrosis seems to contribute to abnormalities in cardiac relaxation and contractility and to the development of heart failure, a common finding in AF. Given that AF and heart failure frequently coexist and that both conditions affect patient prognosis, a better understanding of the mutual effect of fibrosis in AF and heart failure is of particular interest. In this review paper, we provide an overview of the general mechanisms of cardiac fibrosis in AF, differences between fibrotic processes in atria and ventricles, and the clinical and prognostic significance of cardiac fibrosis in AF. (J Am Coll Cardiol 2015;66:943–59) © 2015 by the American College of Cardiology Foundation.

T

he mechanisms of atrial fibrillation (AF) are

Despite a large body of experimental and clinical

complex and associated with structural and

evidence supporting the role of atrial fibrosis in AF,

electrical remodeling in the atria and ventric-

data on fibrotic processes in the ventricles of patients

ular myocardium. The key electrophysiological mech-

with AF are limited. The available data indicate that

anisms of AF include: 1) focal firing due to triggered

ventricular fibrosis may be at least partly responsible

activity (early and delayed afterdepolarizations);

for the impaired cardiac relaxation and contractility

2) multiple re-entries due to shortening of the action

seen in many patients with AF. Cardiac fibrosis may

potential; and 3) heterogeneity of impulse conduc-

be implicated in complex interactions between AF

tion caused by atrial fibrosis. Development and

and heart failure (HF), each of which can be the cause

progression of atrial fibrosis are the hallmark of struc-

and the consequence of the other. Given that AF and

tural remodeling in AF and are considered to be the

HF coexist at high frequency, and their clear prog-

substrate

Advanced atrial

nostic significance (e.g., increased risk of hospitali-

fibrosis is associated with more frequent paroxysms

zation or death related to HF deterioration), a better

for

AF perpetuation.

of AF, transformation of the arrhythmia into a perma-

understanding of the role of cardiac fibrosis in the

nent type, and reduced effectiveness of antiar-

pathogenesis of AF and its complications is impor-

rhythmic drug therapy (1,2).

tant (3). The present review focuses on general

From the *University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Birmingham, United Kingdom; yGrodno State Medical University, Grodno, Belarus; and the zThrombosis Research Unit, Department of Clinical Medicine, Aalborg University, Aalborg, Denmark. Dr. Dzeshka was supported by a European Heart Rhythm Association Academic Fellowship Programme. Dr. Lip has served as a consultant for Bayer, Astellas, Merck, Sanofi, Bristol-Myers Squibb/Pfizer, Biotronik, Medtronic, Portola, Boehringer Ingelheim, Microlife, and Daiichi-Sankyo; and has been on the speakers bureau for Bayer, Bristol-Myers Squibb/Pfizer, Medtronic, Boehringer Ingelheim, Microlife, and Daiichi-Sankyo. Drs. Snezhitskiy and Shantsila have reported that they have no relationships relevant to the contents of this paper to disclose. Listen to this manuscript’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster. Manuscript received May 15, 2015; revised manuscript received June 18, 2015, accepted June 22, 2015.

944

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

ABBREVIATIONS

mechanisms of cardiac fibrosis in AF, differ-

The physiological functions of fibroblasts extend

AND ACRONYMS

ences between fibrotic processes in the atria

beyond metabolism of the ECM. Tight connections

AF = atrial fibrillation CHADS2 = congestive heart failure, hypertension, age

and ventricles, and the clinical and prog-

between the fibroblasts, fibers of the ECM, and other

nostic effect of cardiac fibrosis in AF.

cellular

MECHANISMS OF CARDIAC FIBROSIS

‡75 years, diabetes mellitus, and stroke/transient ischemic

components

form

a

multidimensional

network that acts as an integral sensor of dynamic changes in the various mechanical, chemical, and electrical stimuli in the myocardium. In response to

attack

Progressive accumulation of fibrotic tissue in

CMR = cardiac magnetic

myocardium is 1 of the major components of

resonance

turnover and regulates cardiomyocyte hypertrophy

cardiac remodeling. Formation and redistri-

CTGF = connective tissue

bution of connective tissue fibers modulate

and, to a smaller extent, cardiomyocyte proliferation;

growth factor

myocardial geometry to adapt to new con-

DE-CMR = delayed gadolinium

ditions of (patho)physiological functioning

enhancement cardiac magnetic resonance

and to prevent or minimize the effects of

ECM = extracellular matrix

new mechanical, chemical, and electrical

these stimuli, this complex system adjusts ECM

it also triggers activation of fibrotic and inflammatory pathways. Of note, cardiac fibroblasts exhibit various phenotypes depending on the surrounding microenvironment (10). Cardiac

fibroblasts

might

also

contribute

to

stimuli. This adaptation process involves

electrical remodeling in AF due to their different

both the cellular components of the myo-

electrophysiological properties compared with the

cardium and the extracellular matrix (ECM),

surrounding cardiomyocytes. Fibroblasts are essen-

metalloproteinase

an acellular component of the heart, containing a variety of fibers, with collagen

tially nonexcitable cells but can transfer currents

NADPH = nicotinamide adenine dinucleotide

predominant (4).

HF = heart failure miR = microribonucleic acid MMP = matrix

phosphate

between cardiomyocytes via connexins in vitro. This action may result in heterogeneity of current con-

Excessive ECM production in adults is

duction, shortening of action potentials, depolariza-

PDGF = platelet-derived

commonly associated with the pathogenesis

growth factor

tion of resting cardiomyocytes, and induction of

of cardiovascular diseases, resulting in ab-

spontaneous phase 4 depolarization (11). Conse-

normalities of cardiac contraction and relax-

quently, fibroblasts might be directly involved in the

TGF-b1 = transforming growth factor beta-1

ation, and thus inevitably leading to HF (5).

occurrence and perpetuation of re-entry, although

Although collagen deposition in the healthy heart is

further research is needed to provide in vivo evidence

restricted to maintenance of heart architecture, dur-

for this claim. Interestingly, computer modeling

ing progression of various cardiac disorders, the

found proliferation of myofibroblasts in AF and their

collagen network undergoes quantitative and quali-

electrical interaction with cardiomyocytes to be suf-

tative changes leading to excessive accumulation of

ficient for re-entry formation, even in the absence of

collagen, either in regions of cardiomyocyte loss (e.g.,

fibrosis (12).

in myocardial infarction, reparative fibrosis) or diffusely, in areas of the myocardium not involved in the focal injury (e.g., in dilated cardiomyopathy, reactive fibrosis) (4,5).

Myofibroblasts are cells that play a particularly significant role in cardiac fibrosis. They are derived from cardiac fibroblasts but have an w2-fold higher capacity to synthesize collagen. Compared with car-

CARDIAC FIBROBLASTS AND MYOFIBROBLASTS. Both

diac fibroblasts, myofibroblasts do not appear in

cellular and extracellular components take part in the

healthy myocardium, are more responsive to proin-

remodeling process. Cardiac fibroblasts play a pivotal

flammatory and profibrotic stimuli, and are capable

role in formation of the ECM. They are numerous

of synthesizing a large variety of cytokines and

within the myocardium and can account for up to

chemokines (13). Importantly, myofibroblasts contain

60% of cells in cardiac muscle (6). Thus, cardiac fi-

alpha-smooth muscle actin and adhesion complexes

broblasts outnumber even cardiomyocytes, although

(fibronexus). The latter binds myofibroblast internal

the latter cells largely determine total myocardial

microfilaments to ECM proteins that help to provide

mass.

contractile force to the surrounding ECM.

The population of cardiac fibroblasts in healthy

A range of growth factors, cytokines, and hor-

adult hearts is maintained at a relatively low

mones, as well as mechanical stretch and hypoxia,

level and is predominantly composed of resident

regulate ECM turnover and cardiac fibroblast activity.

fibroblasts and cells undergoing an epithelial-to-

These factors determine fibroblast gene expression,

mesenchymal transition. In pathological conditions,

their differentiation, and the level of collagen

fibroblasts dramatically increase in number via dif-

synthesis.

ferentiation from several cell lineages, including

TRANSFORMING GROWTH FACTOR-b1 SIGNALING. Among

monocytes, endothelial cells, bone marrow–circu-

the numerous regulatory factors, angiotensin II and

lating progenitor cells, and pericytes (7–9).

transforming growth factor beta-1 (TGF-b 1) (Figure 1)

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

are the most potent stimulators of collagen synthesis by cardiac fibroblasts (14,15). TGF-b1 produces its ef-

Dzeshka et al. Cardiac Fibrosis in Patients With Atrial Fibrillation

F I G U R E 1 Schematic Overview of the TGF- b1 –Signaling Pathway in Cardiac Fibrosis

fects via binding to the dimerized TGF- b 1 receptor (which consists of 2 receptors, Tb RI and T bRII) in the extracellular space. Ligand-receptor binding results in a cascade of phosphorylation reactions during which inactive Smad proteins 2, 3, and 4 form the Smad complex (16). The Smad complex then translocates to the nuclei of target cells, where it regulates expression of genes involved in fibrogenesis (e.g., connective tissue growth factor [CTGF] and periostin) via appropriate regulatory regions (17,18). This action results in production of the so-called matricellular protein, a profibrotic protein secreted into the ECM. The matricellular protein modulates intercellular and cell–matrix interactions that further stimulate ECM protein synthesis but are not directly involved in ECM structure and mechanical organization or differentiation of cardiac fibroblasts into myofibroblasts (19). TGF- b–activated kinase 1, an alternative to the Smad pathway for TGF- b1–induced fibrosis, is a member of the mitogen-activated protein kinase family (20). Importantly, apart from activation of fibroblast and collagen synthesis, TGF-b 1 can also induce apoptosis of cardiomyocytes (21). Of note, angiotensin II is not able to induce cardiac hypertrophy and fibrosis in the absence of TGF-b1, but it up-regulates TGF- b1 synthesis, Smad2 phosphorylation, and nuclear translocation of the Smad complex and increases Smad deoxyribonucleic acid— binding activity. TGF- b 1, in turn, can directly stimulate expression of angiotensin II type 1 receptor (22). Angiotensin II also predisposes to fibrosis by promoting expression of profibrotic factors, such as endothelin-1. In conjunction with aldosterone, angiotensin II promotes oxidative stress (i.e., excess production of reactive oxygen species) and inflammation, mainly by activation of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (23,24). MATRIX METALLOPROTEINASES AND TISSUE INHIBITORS OF MATRIX METALLOPROTEINASES. The remodeling

and maintenance process of the extracellular space includes not only synthesis but also coordinated degradation of the ECM proteins. Matrix metalloproteinases (MMPs) and their tissue inhibitors, synthesized by cardiomyocytes and cardiac fibro-

COL1A1 ¼ geneencodingalpha-1 typeI collagen; COL1A2 ¼ gene encoding alpha-2type Icollagen;

blasts, are intimately involved in the maintenance of

COL3A1 ¼ gene encoding a1 type III collagen; CTGF ¼ connective tissue growth factor; ECM ¼

ECM homeostasis (25). Indeed, MMP expression in-

extracellular matrix; ERK ¼ extracellular signal-regulated kinase; JNK ¼ c-jun N-terminal kinase;

creases in a time-dependent manner with left ventricular dysfunction and dilation (26). Overexpression of MMP1 has been shown to cause compensatory

LOX ¼ lysyl oxidase; LTBP ¼ latent transforming growth factor beta-1 binding protein; MMP ¼ matrix metalloproteinase; p38 ¼ protein 38 (member of MAPK, mitogen-activated protein kinases); SAPK ¼ stress-activated protein kinase; Smad ¼ a transcription factor (as discussed in Table 1); TAK1 ¼ transforming growth factor beta-1 activated kinase 1; TbRI ¼ type I transforming

hypertrophy and increased collagen concentration

growth factor beta-1 receptor; TbRII ¼ type II transforming growth factor beta-1 receptor;

within the myocardium. In contrast, targeted deletion

TGF-b1 ¼ transforming growth factor beta-1; TIMP ¼ tissue inhibitor of matrix metalloproteinase.

945

946

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

of MMP2 results in amelioration of left ventricular

Several miRs are involved in fibrogenesis. miR-133

remodeling (27). Not surprisingly, MMP activity in-

and miR-30 regulate cardiac fibrosis by repressing

creases in line with TGF-b 1 expression within the

CTGF expression. They were found to be down-

myocardium and correlates with the level of inflam-

regulated in left ventricular hypertrophy associated

mation and oxidative stress (28).

with

Furthermore, collagen and matrix fragments pro-

increased

CTGF

expression

(33).

miR-133

knockout mice developed advanced fibrosis and HF

duced by the action of MMP1 themselves form

with predisposition to sudden death (34). In contrast,

bioactive molecules, so-called matrikines, and release

miR-133

ECM-embedded

collagen synthesis by fibroblasts, reduced myocardial

proinflammatory

and

profibrotic

overexpression

resulted

in

decreased

factors. They promote fibroblast activation and tran-

fibrosis, and apoptosis (33,35). miR-21 is involved in

sition to a myofibroblast phenotype, and they effec-

up-regulation of 1 of the profibrotic pathways (ERK)

tively stimulate connective tissue synthesis by

and promotes MMP2 expression (36,37). Interestingly,

serving as ligands of leukocyte integrins and other

it also produces protective effects, including defense

cell-activating receptors (25,29). The latter explains

against oxidative stress, inhibition of proapoptotic

the progression of fibrosis when MMP activity is high,

factors, and increased expression of antiapoptotic

despite the primary MMP function being directed

genes (38). Finally, miR-29 is associated with depo-

toward matrix degradation. MMP activity is regulated

sition of collagen types I and III. Up-regulation of

via tissue inhibitors of MMPs and reversion-inducing

miR-29 leads to down-regulation of these proteins

cysteine-rich

and vice versa (39).

protein

with

Kazal

motifs,

over-

expression of which has been found to blunt angio-

INFLAMMATION. AF is common in patients with

tensin

overt inflammatory states of cardiac and noncardiac

II–induced

MMP

activation

and

cardiac

fibroblast migration (30). REGULATORY

ROLE

OF

location (e.g., myocarditis, pericarditis, pneumonia, MICRORIBONUCLEIC

ACIDS.

and inflammatory bowel disease), but low-grade

Nuclear microribonucleic acids (miRs) play important

subclinical inflammation (e.g., in coronary heart

regulatory roles in cardiac remodeling (31). They

disease) also contributes to pathogenesis of the

are referred to as endogenous, single-stranded, short

arrhythmia (Figure 2) (40). Regardless of whether

(w22

acids.

AF is a cause or consequence of the inflammatory

miRs degrade or inhibit the translation of their

process, it is related to oxidative stress perpetuated

target messenger ribonucleic acids at the post-

by myocardial infiltration with inflammatory cells

transcriptional level, thus regulating gene expres-

(e.g., macrophages), which is accompanied by release

sion (32).

of reactive oxygen species. Inflammation is further

nucleotides),

noncoding

ribonucleic

F I G U R E 2 Involvement of Inflammation Pathways in Cardiac Fibrosis

Tissue injury (AF, HF)

Activated inflammatory cells (neutrophils, lymphocytes, monocytes, resident macrophages, mast cells, platelets)

Low grade inflammatory response

IL-1, IL-6, TNFα α, MPO, PDGF

Transcriptional factors (e.g. NF-kB)

Gene expression of inflammatory cytokines

Maintenance of inflammation, RAAS activation, oxidative stress, TGF-β1 signaling

Apoptosis of cardiomyocytes, fibroblast recruitment, ECM deposition

AF ¼ atrial fibrillation; HF ¼ heart failure; IL ¼ interleukin; MPO ¼ myeloperoxidase; NF-kB ¼ nuclear factor kappa B; PDGF ¼ platelet-derived growth factor; RAAS ¼ renin-angiotensin-aldosterone system; TGF-b1 ¼ transforming growth factor b1; TNF ¼ tumor necrosis factor.

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

exacerbated by activation of the renin-angiotensin-

volume fraction in the myocardium and by result of

aldosterone

imaging (44).

system,

followed

by

activation

of

consequently

Age-related cardiac fibrosis reflects multiple pro-

trigger TGF- b1 signaling and structural and electrical

cesses that accompany cardiac senescence, chronic

remodeling (41). Various inflammatory cytokines and

activation of the renin-angiotensin-aldosterone axis,

chemokines, such as interleukin-1 and -6, tumor ne-

excessive beta-adrenergic and endothelin signaling,

crosis factor-alpha, and monocyte chemoattractant

activation of the TGF-b1 pathway, disruption of

protein 1, are up-regulated in AF and linked to pro-

intracellular calcium homeostasis, cardiomyocyte

gression from paroxysmal to chronic AF and AF

apoptosis, recruitment of mononuclear cells and

recurrence post-cardioversion (40).

fibroblast progenitors, and down-regulation of the

NADPH

oxidase.

These

processes

Inflammation plays a particular role in postoperative AF (e.g., after coronary artery bypass graft,

mitochondrial nicotinamide adenine dinucleotide– dependent deacetylase sirtuin-1 (45).

valvular replacement surgery) and post–catheter

Increased generation of reactive oxygen species

ablation. In a recent meta-analysis of 925 post-

and diminished antioxidant capacity are major con-

operative patients, serum C-reactive protein was a

tributors

potent predictor of new-onset AF (42). Similarly, a

(Figure 3). Oxidative molecules derive from oxidative

meta-analysis of 7 studies of post-ablation patients

phosphorylation processes in mitochondria, increased

confirmed the predictive role of C-reactive protein for

NADPH oxidase activity, uncoupled nitric oxide syn-

AF recurrence (43).

thase function, lipid oxidation within peroxisomes,

to

age-related

myocardial

remodeling

and up-regulation of cyclooxygenases and xanthine

AGING AND CARDIAC FIBROSIS

oxidase (46). Chronic oxidative stress leads to the persistence of low-grade inflammation, thus further

The prevalence of AF significantly increases in the

accelerating cardiac fibrosis. Among the important

elderly. Cardiac aging is a complex process, featuring

regulators of aging-related processes is miR-34a, with

progressive decline in heart functions and ventricular

phosphatase 1 nuclear–targeting subunit (PNUTS),

and atrial remodeling. This process includes re-

also known as PPP1R10, as the target. Aging-induced

duction in cardiomyocyte numbers, hypertrophy of

expression of miR-34a and inhibition of PNUTS is

the remaining cardiomyocytes, alteration of myofi-

associated with telomere shortening, deoxyribonu-

brillar orientation, proliferation of cardiac fibroblasts,

cleic acid damage, cardiomyocyte apoptosis, and

and collagen deposition. Progressive fibrosis is a

impaired functional recovery after ischemic injury

hallmark of the aging heart, as confirmed in animal

(47). Hence, profibrotic mechanisms involved in AF

and human studies showing an increased collagen

pathogenesis are clearly enhanced by aging.

F I G U R E 3 Oxidative Stress in Cardiac Fibrosis

Fibrotic remodeling Target genes expression

Tissue injury (AF, HF), RAAS activation

NADPH oxidase, xanthine oxidoreductase, uncoupled NOS, electron transport chain in the mitochondria

Intracellular accumulation of ROS (oxidative stress)

Signaling pathways (MAPK, ERK, JNK, p38)

Transcriptional factors (e.g., NF-kB)

ERK ¼ extracellular signal-regulated kinase; JNK ¼ c-Jun N-terminal kinase; MAPK ¼ mitogen-activated protein kinase; NADPH ¼ reduced nicotinamide-adenine dinucleotide phosphate; NOS ¼ nitric oxide synthase; p38 ¼ protein 38; ROS ¼ reactive oxygen species; other abbreviations as in Figure 2.

947

948

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

CARDIOMYOCYTE–CARDIAC

molecular pathways (mediated by angiotensin II,

FIBROBLAST COMMUNICATION

TGF- b1, endothelin, and cytokines). However, the response to signaling may vary depending on cell

Close interaction between cardiomyocytes and car-

type: hypertrophy and reduced cell survival of car-

diac (myo)fibroblasts is essential for their function.

diomyocytes are promoted by angiotensin II–induced

These interactions are facilitated by multiple para-

release of TGF-b 1 and endothelin-1 from fibroblasts,

crine signals, including those predisposing to fibrosis

whereas angiotensin II was also found to trigger

(Central Illustration). Cardiomyocytes, cardiac fibro-

release of TGF-b1 and endothelin-1 from cardio-

blasts, and myofibroblasts share many common

myocytes and to stimulate fibroblast proliferation and differentiation to the myofibroblast phenotype

C EN T RA L IL LUSTR AT I ON

Cardiomyocytes, Cardiac Fibroblasts, and Myofibroblasts Crosstalk and Paracrine Factors Mediating Profibrotic and Antifibrotic Effects

and synthesis of ECM components (48). Greater proliferation of fibroblasts was observed around cardiomyocytes expressing angiotensin II receptor type 1 receptors, compared with cells with the angiotensin II receptor type 1 gene knocked out (49). There are also differences between cardiomyocytes and cardiofibroblasts in receptor density and receptor kinase activity, which may interfere with the final effects of effector molecules. For example, fibroblasts are known to carry more angiotensin II receptors than cardiomyocytes (49). Multiple other regulatory substances are implicated in the interplay between fibroblasts and cardiomyocytes (e.g., fibroblast growth factor 2, interleukins, natriuretic peptides, and miRs) (48). Some stimuli are attributed predominantly to fibroblasts (e.g., platelet-derived growth factor [PDGF], fibroblast growth factor 2, and activation by mechanical stretching), whereas abnormalities in calcium handling are largely seen in cardiomyocytes (50). Cardiac remodeling requires fibrosis as an essential component and is a complex process that also incorporates multiple other pathways, such as hypoxia signaling, the osteoprotegerin/RANK/RANKL axis, and Ca2þ signaling. A detailed description of these pathways is beyond the scope of the current review. In summary, the ECM is a macromolecular, metabolically active, dynamic network of fibers (predominantly collagen) and cells (predominantly cardiac fibroblasts with a capacity to differentiate into myofibroblasts) that is essential for normal heart function. Cells surrounding and infiltrating the ECM are linked to the fibrillar component and hence are capable of responding to mechanical stretch and stress, as well as to a variety of autocrine and paracrine stimuli by change in their proliferation, migration, and intensity of collagen synthesis. These processes may have an unfavorable role in cardiac remodeling and the

Dzeshka, M.S. et al. J Am Coll Cardiol. 2015; 66(8):943–59.

Aldo ¼ aldosterone; Ang II ¼ angiotensin II; CTGF ¼ connective tissue growth factor; ECM ¼

pathogenesis of cardiovascular diseases.

ATRIAL FIBROSIS IN AF

extracellular matrix; ET ¼ endothelin; FGF ¼ fibroblast growth factor; IL ¼ interleukin; miR ¼ microribonucleic acid; PDGF ¼ platelet-derived growth factor; ROS ¼ reactive

A variety of signaling systems are involved in the

oxygen species; TGF ¼ transforming growth factor; TNF ¼ tumor necrosis factor.

promotion

of

atrial

fibrosis,

as

evidenced

by

numerous human and animal data (Tables 1 and 2).

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

Atrial fibrosis may develop as part of AF-related

been well documented in several animal models of

structural remodeling, as well as a consequence of

AF, cardiac fibrosis is unlikely to be the sole mecha-

other cardiovascular diseases that result in atrial

nism of HF, including HF with preserved ejection

overload and stretch. Conditions associated with

fraction. For instance, predominant electrical re-

atrial fibrosis include hypertension, valvular heart

modeling with minimal (if any) evidence of atrial

disease, and HF, which cause broadly similar histo-

fibrosis and preservation of ventricular contractility

logical changes in the atrial myocardium (51). How-

was seen in a sheep model of prolonged persistent AF

ever, the precise causality between AF and atrial

induced by intermittent atrial tachypacing, but there

fibrosis may be difficult to establish.

was no significant tachycardia during the observation

Experimental data have also yielded controversial

period (60). More recently, Martins et al. (61) reported

results. For example, some models of atrial tachyar-

that long-term AF with duration up to 1 year and

rhythmia

dilation

transition to persistent arrhythmia was accompanied

with changes in atrial architecture and myocyte

by progressively increasing atrial dilation, mitral

characteristics (e.g., loss of myofibrils; accumulation

valve regurgitation, myocyte hypertrophy, and at-

of glycogen; changes in mitochondrial number,

rial fibrosis, with no evidence of left ventricular

shape, and size; fragmentation of sarcoplasmic retic-

dysfunction.

demonstrated

marked

biatrial

ulum; dispersion of nuclear chromatin) while the

Thus, atrial remodeling is the mainstay of AF

interstitial space remained unaltered, with no evi-

initiation and perpetuation. Atrial structural and

dence of increased connective tissue content (52). In

functional changes may develop as a result of un-

contrast, more recent studies of rapid atrial pacing

derlying cardiac conditions, pathological systemic

demonstrated up-regulation of potent profibrotic

processes, or AF itself. Atrial remodeling also com-

factors, such as angiotensin II and TGF- b1 (53), and

monly occurs as a part of age-related processes.

increased collagen content in the atrial interstitium

However, the relationship between the course of AF

(54). The discrepancy might be attributable to the

and atrial fibrosis is complex and nonlinear, meaning

time required for development of detectable fibrosis

that higher levels of collagen deposition within the

after initiation of profibrotic pathways. For example,

atria do not always cause more frequent paroxysms

in a mouse model of HF, there were no signs of his-

of the arrhythmia and its progression toward the

tological fibrosis in the left atrium at 8 weeks, despite

persistent or permanent type. A plethora of mecha-

increased expression of genes related to fibrosis (55).

nisms (i.e., hemodynamic alterations, mechanical

This discrepancy may also suggest that despite the

stretching, changes in hormones, growth factors, and

activation of profibrotic genes, other, poorly under-

proinflammatory cytokines) modulates the severity

stood mechanisms of inhibition of fibrosis exist.

of atrial fibrosis. A large body of experimental and

Background cardiovascular disease causing HF can be associated with more advanced atrial changes. For

clinical data has already provided insights into the key mechanisms of atrial fibrosis.

example, a model produced by a combination of rapid atrial pacing with mitral regurgitation inevitably

EXTRACARDIAC AND GENETIC FACTORS

resulted in intercellular space expansion in the left

CONTRIBUTING TO ATRIAL FIBROSIS

atrium and AF (56). This observation is consistent with an HF model of AF caused by ventricular

Multiple noncardiac factors predispose to fibrosis in

tachypacing, in which connective tissue contained

AF, including obesity, metabolic syndrome, use of

increased numbers of fibroblasts, more collagen, and

toxic substances, athlete’s heart, obstructive sleep

signs of degeneration and necrosis, compared with an

apnea, systemic inflammation, and thyrotoxicosis. All

atrial pacing model (57). Cardin et al. (58) noted that

of these factors ultimately affect the myocardium

ventricular tachypacing led to an w10-fold over-

(62). Recently, diabetes, a disease associated specif-

expression of collagen messenger ribonucleic acid in

ically with cardiomyopathy and excessive cardiac

atrial cardiomyocytes as early as 24 h and progressed

fibrosis, was shown to triple the risk of AF in obese

further at 2 weeks, compared with atrial pacing. Of

subjects (63). Obesity leads to electrical and structural

note, 8 collagen genes were up-regulated >10-fold,

atrial remodeling and is associated with diastolic

fibrillin was up-regulated 8-fold, and MMP2 was up-

ventricular impairment, atrial dilation, and myocar-

regulated 4.5-fold at 2 weeks; however, there were

dial lipidosis (64). Obesity in AF is related to delay

no changes in their expression at 24 h. In addition,

and significant heterogeneity in atrial conduction,

TGF- b 1 levels in failing hearts in animals appeared to

atrial

be higher than in nonfailing hearts (59). Although

fibrosis. Pathways underlying these changes include

development of atrial fibrosis associated with HF has

activation of TGF-b1 signaling, oxidative stress, and

inflammatory

infiltration,

and

interstitial

949

950

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

T A B L E 1 Studies on Mechanisms of Atrial Fibrosis in AF in Humans

n (AF/SR)

Sample Tested

LVEF, % (AF/Controls)

Adam et al., 2010 (105)

5/5

LAA

61  6/59  6

Adam et al., 2012 (106)

5/5

LAA

61  6/59  6

Cao et al., 2013 (107)

48/24

RAA

63  5 (paroxysmal), 64  5 (persistent)/ 64  3

Dawson et al., 2013 (108)

17/30* 17/19

Plasma RAA

60  2/69  1

Gramley et al., 2007 (109)

42/104

RAA

48  12

Gramley et al., 2010 (68)

42/116

RAA

50  11/47  12

Gramley et al., 2010 (110)

61/102

RAA

48  11

Kallergis et al., 2008 (111)

70/20

Serum

60  4 (paroxysmal), 56  9 (persistent)/ 60  5

Ko et al., 2011 (112)

10/10

RAA

53  15/44  17

Li et al., 2013 (113)

28/12

RA

NA

Mayyas et al., 2010 (114)

32/21

LAA

51  2/53  3

Nishi et al., 2013 (115)

16/13

RA

70  8 (unsuccessful Maze), 53  15 (successful Maze)/ 62.0  9

Okumura et al., 2011 (116)

50/0

Serum

NA

Polyakova et al., 2008 (117)

24/24

RA, RAA

46  10/46  13

[collagen content, MMP2, MMP9, TIMP1, TIMP2, RECK, TGF-b1, Smad2 and phSmad2

Qu et al., 2009 (118)

20/20

RA

51  15/63  14.63

[collagen content, TNFa, IL-6 and NFkB activity

Rahmutula et al., 2013 (59)

17/NA

RA

NA

[TGb1 and TGF-b1 signaling–related proteins (e.g., phSmad2, Smad6, Ang II)

Richter et al., 2011 (119)

30/0

Serum

62  2

Rudolph et al., 2010 (120)

34/35

Plasma, RAA

49  89/52  9

Swartz et al., 2012 (121)

18/36

LAA, RAA, serum

49  12/51  8

[collagen content, collagen type I, III, TGF-b1, Ang II (mRNA) [PICP, PIIINP

Collagen content with PICP

Wang et al., 2015 (122)

30/17

LAA

63  7/70  4

[miR-146b-5p, MMP9, collagen content; YTIMP4

miR-146b-5p with TIMP4 and collagen content

First Author, Year (Ref. #)

Wilhelm et al., 2006 (123)

30/20

RAA

NA

Wu et al., 2013 (124)

200/0

Plasma

53  10/57  6†

Results (AF vs. Controls)

[collagen, CTGF, NADPHox, Rac1, N-cadherin, connexin 43, Ang II [miR-21 [OPG, RANKL, RANK, RANKL/OPG ratio

Observed Associations

NA miR-21 with collagen, CTGF, Rac1, LOX, Ang II (In AF) OPG, RANKL, RANK, RANKL/OPG ratio with collagen types I and III

YmiR-29b YmiR-29b in chronic AF

NA

[collagen content, activity 4MMP2, MMP9 (mRNA and protein levels), YPAI, TIMP1 and 2 (mRNA) with [ duration of AF

NA

[collagen content, HIF-1a, HIF-2a, VEGF, KDR, pKDR, and microvessel density

NA

[collagen content, early [ and later Y responsiveness to TGF-b1 with [ duration of AF: initially [TGF-b1 (mRNA and protein), TbRII, phSmad2, Smad4 (protein) followed by a YTbRI phSmad2 (protein) and [Smad7 (protein)

NA

[CITP, CICP, TIMP1

NA

[collagen content, CTGF (protein and mRNA)

NA

[collagen content, TGF-b1, Smad3, and CTGF

TGF-b1, CTGF (mRNA and protein) with collagen content, TGF-b1, CTGF

[ET-1 4ETAR or ETBR

ET-1 with LA size, AF persistence

[miR-21, miR-23b, miR-199b, miR-208b

miR-21 with collagen content

YhsCRP, IL-6, ANP, BNP [MMP2, TIMP2, CITP during follow-up

MMP2 with AF recurrence

[MMP9, TGF-b1, PIIINP after ablation

[MPO

4collagen content NA

NA

NFkB activity with TNFa, IL-6, and collagen content NA

PIIINP with AF recurrence; MMP9, TGF-b1 with ablationinduced LA volume reduction; MMP9 with RF energy on ablation NA

NA TGF-b1 with AF recurrence Continued on the next page

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

951

Cardiac Fibrosis in Patients With Atrial Fibrillation

T A B L E 1 Continued

n (AF/SR)

Sample Tested

LVEF, % (AF/Controls)

83/52†

RAA

62  6/30  5

Xie et al., 2013 (126)

22/15

LA

55/60

Xu et al., 2009 (127)

27/18

LA, RA

54  9/59  12

First Author, Year (Ref. #)

Xi et al., 2013 (125)

Results (AF vs. Controls)

Observed Associations

4RANK, RANKL

RANK, RANKL, RANKL/OPG ratio with collagen content

[fibrocytes, collagen I and alpha-SMA

Fibrocytes with collagen content, LA volume index

[collagen type I, MMP1 and MMP9 (mRNA) YTIMP1 (mRNA)

Collagen type I (mRNA) with atrial diameter; MMP1, MMP9 (mRNA) with TIMP1 (mRNA)

*Recurrent AF/nonrecurrent AF. †Chronic AF/AF with subsequent SR recovery. [ ¼ increased; Y ¼ reduced; 4 ¼ not changed; AF ¼ atrial fibrillation; Ang ¼ angiotensin; ANP ¼ atrial natriuretic peptide; BNP ¼ B-type natriuretic peptide; CICP ¼ collagen type I C-terminal propeptide; CITP ¼ collagen type I C-terminal telopeptide; COL1A1 ¼ gene encoding alpha-1 type I collagen; COL3A1 ¼ gene encoding alpha-1 type III collagen; CTGF ¼ connective tissue growth factor; ET ¼ endothelin; ETAR ¼ type A endothelin-1 receptor; ETBR ¼ type B endothelin-1 receptor; HIF ¼ hypoxia-inducible factor; hsCRP ¼ high-sensitivity C-reactive protein; IL ¼ interleukin; KDR ¼ vascular endothelial growth factor receptor 2; LA ¼ left atrium; LAA ¼ left atrial appendage; LOX ¼ lysyl oxidase; LVEF ¼ left ventricular ejection fraction; miR ¼ microribonucleic acid; MMP ¼ matrix metalloproteinase; MPO ¼ myeloperoxidase; mRNA ¼ messenger ribonucleic acid; NA ¼ not available; NADPHox ¼ nicotinamide adenine dinucleotide phosphate oxidase; NFkB ¼ nuclear factor kappa B; OPG ¼ osteoprotegerin; PAI ¼ plasminogen activation inhibitor; phSmad ¼ phosphorylated Smad; PICP ¼ procollagen type I C-terminal propeptide; PIIINP ¼ procollagen type III N-terminal propeptide; pKDR ¼ phosphorylated KDR; RA ¼ right atrium; RAA ¼ right atrial appendage; Rac1 ¼ Ras-related C3 botulinum toxin substrate 1; RANK ¼ receptor activator of nuclear factor kappa B; RANKL ¼ receptor activator of nuclear factor kappa B ligand; RECK ¼ reversion-inducing cysteine-rich protein with Kazal motifs; SMA ¼ smooth muscle actin; Smad ¼ a transcription factor, named by fusion of the Caenorhabditis elegans Sma and Drosophila Mad (mothers against decapentaplegic) proteins, in reference to its homology to these proteins; SR ¼ sinus rhythm; TbRI ¼ type I transforming growth factor beta-1 receptor; TbRII ¼ type II transforming growth factor beta-1 receptor; TGF-b1 ¼ transforming growth factor beta-1; TIMP ¼ tissue inhibitor of matrix metalloproteinase; TNF ¼ tumor necrosis factor; VEGF ¼ vascular endothelial growth factor.

up-regulation of PDGF and endothelin (65). The

AF development and persistence, and it is likely to

pericardial fat envelope could produce a constrictive

have a genetic background (73,74). One of the genes

effect, thus disturbing cardiac relaxation. Some adi-

involved in atrial development is PITX2. Deficiency of

pokines clearly have profibrotic properties (e.g.,

this gene results in formation of enlarged atria with

activin A, a member of the TGF- b1 superfamily),

thin walls and prominent deficiency in ion channel

whereas AF itself affects adipocyte-related gene

expression (75). AF is also related to polymorphisms

expression, facilitating expansion of cardiac fat (66).

in genes involved in fibrotic pathways, such as those

Obstructive and central sleep apnea are also asso-

responsible for modulation of the synthesis of

ciated with atrial remodeling and AF (67). Animal

interleukin-1 and -6 (2). Nonetheless, there is pres-

models of obstructive sleep apnea showed substantial

ently a lack of robust evidence of target genes directly

connexin 43 down-regulation and altered expression

responsible for AF-related cardiac fibrosis, and

of channel proteins, with net effects of shortening the

further research is warranted.

atrial refractory period, slowing atrial conduction, cardiomyocyte

apoptosis,

and

cardiac

fibrosis.

VENTRICULAR FIBROSIS IN AF

Repeated apnea episodes associated with chronic hypoxia trigger production of strongly profibrotic

Association of AF with ventricular fibrosis is less

hypoxia-inducible factors 1a and 2 a (68). Moreover,

established than for atrial fibrosis. Ventricular fibrotic

such patients have increased angiotensin-converting

changes are more pronounced in AF patients than in

enzyme and interleukin-6 expression, with inhibited

subjects with sinus rhythm, both with magnetic

degradation of atrial collagen via MMP2 (69). In

resonance or ultrasound imaging (76,77). In addition,

addition, sleep apnea leads to myocardial hypertro-

more extensive changes were found in patients with

phy and diastolic dysfunction, thus further potenti-

permanent or persistent arrhythmia compared with

ating development of HF in the presence of AF

those with paroxysmal AF (76–78). These reports were

(70,71). Multiple other factors contribute to atrial

confirmed by use of animal data implicating ventric-

remodeling and chronic low-grade inflammation with

ular fibrosis in cardiac remodeling and rate control in

increased propensity to AF (e.g., chronic kidney dis-

AF (79). Avitall et al. (80) observed more extensive

ease, diabetes) (11).

ventricular fibrosis in AF when ventricles were not

AF has a genetic predisposition. The genetic background of AF is complex, involving multiple path-

protected from a high atrial rate with atrioventricular node ablation than in ablated animals.

ways. In “lone” AF (i.e., in the absence of apparent

Atrial and ventricular fibrosis in AF is likely to

cardiovascular disease), histological and imaging ev-

share many common mechanisms, although the

idence suggests a similar extent of atrial fibrosis as in

extent may vary between the 2 parts of the heart. In

AF patients with structural heart disease (72). A spe-

transgenic mice with TGF- b 1 overexpression, TGF- b 1

cific fibrotic atrial cardiomyopathy has been sug-

up-regulation was more pronounced in atria than in

gested as an underlying condition predisposing to

ventricles. High TGF- b1 levels were associated with

952

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

T A B L E 2 Studies on Mechanisms of Atrial Fibrosis in AF in Animal Experiments

First Author, Year (Ref. #)

n (Experiment/Control)

Abed et al., 2013 (65)

30

Adam et al., 2012 (106)

NA

Cardin et al., 2012 (128)

NA

Samples Tested

LA, LAA, RA, and RAA LA Myocardium

Model

Results (Experiment Vs. Controls)

High-calorie diet

[AF inducibility, LA volume, collagen content, inflammatory infiltrates, lipidosis, ET-1, ETAR, ETBR, TGF-b1, PDGF with [ adiposity

Tx, Rac1 overexpression

[collagen content, miR-21, spontaneous AF

Coronary artery ligation

[miR-21, LA dilation and collagen content, AF inducibility

57/37

LA

Ventricular tachypacing; Tx, miR-29b knockout

YmiR-29b, miR-133a, miR-133b; [collagen content [collagen content, COL1A1 (mRNA)

8/8

LA

Atrial tachypacing

[Ang II, TGF-b1, phSmad2/3, Arkadia, hydroxyproline expression; YSmad7 expression

Ko et al., 2011 (112)

6/6

LA, RA

Atrial tachypacing

[Ang II, CTGF (protein and mRNA)

Li et al., 2012 (54)

21/21

LA

Atrial tachypacing

[collagen content, chronic inflammation; YmiR-133 and miR-30

Rahmutula et al., 2013 (59)

15/15

Tx, TGF-b1 overexpression

[LA fibrosis, AF inducibility, TGF-b1 signaling–related genes in atria (TIMP, MMP, collagen, Smad2 or 3, TbRI and II)

Rudolph et al., 2010 (120)

NA

MPO knockout, angiotensin II pre-treatment

Ycollagen content, MPO, MMP2, MMP9, and AF susceptibility

Dawson et al., 2013 (108)

He et al., 2011 (53)

Myocardium LA

Saba et al., 2005 (129)

32/37

Myocardium

Tx, TNF alpha overexpression

[collagen content, AF inducibility

Verheule et al., 2004 (130)

30/30

Myocardium

Tx, TGF-b1 overexpression

[LA fibrosis, AF inducibility

PDGF ¼ platelet-derived growth factor; Tx ¼ transgenic; other abbreviations as in Table 1.

enhanced expression of only 2 profibrotic genes in

TGF- b1 in ventricles versus atria (Table 3). Impor-

ventricles but of 80 genes in atria. Interestingly, T b RI,

tantly, the processes described earlier were reported

T bRII, and Smad protein levels were similar in ven-

in intact ventricles. When TGF-b 1 overexpression was

tricles and atria, but Smad2 phosphorylation was

combined with other pathophysiological stimuli (e.g.,

increased only in atria, making them prone to devel-

those seen in HF), TGF-b1 –mediated profibrotic ven-

opment of selective fibrosis (59). Several mechanisms

tricular effects appeared to be enhanced, although

have been suggested for the different effects of

atrial fibrosis still predominated. In addition, atrial fibroblasts are more susceptible to PDGF, angiotensin

T A B L E 3 Suggested Mechanisms of Predisposition of

II, and endothelin-1, indicating that their activity

Atrial Versus Ventricular Myocardium to Fibrosis

is pre-modulated by local atrial factors (81). Hemo-

Gene expression

dynamic changes (e.g., stretch) have a differential

Higher expression of genes encoding extracellular matrix (e.g., fibronectin, laminin, fibulin), cell signaling (PDGF, PDGF receptor, angiopoietin, VEGF), in fibroblasts Signaling pathways Greater signaling via canonical TGF-b1 pathway including enhanced receptor binding, receptor-kinase activity, Smad2/3 phosphorylation, reduced expression of the inhibitory Smad7 in atrial myocytes with further stimulation of atrial fibroblasts Higher level of endogenous AT1 receptor and greater enhancement of receptor level following pathological influences Differential expression of adapter/scaffolding proteins (b-arrestin-1, G-proteins) involved in the AT1 receptor–dependent aldosterone synthesis and secretion Higher expression of PAI-1 Cellular proliferation Higher myofibroblast density in healthy and diseased hearts with faster cell surface area being increased, distinct morphology at confluence, and greater alpha-SMA and vimentin expression Higher proportion of fibroblasts in mitotic phases and displaying enhanced gene expression of fibroblast-selective markers Greater proliferation response of atrial fibroblasts for a range of growth factors (e.g., PDGF, angiotensin II, ET-1, TGF-b1)

effect, and the precise biochemical processes acting on fibroblast activity in ventricles and atria remain unclear at present. In summary, the cardiac profibrotic microenvironment in AF is unlikely to be strictly limited to the atria, and the ventricular myocardium is also likely to be affected. Even within the atrial myocardium, fibrotic changes take years to become detectable with the diagnostic methods currently available. With the considerably higher myocardial thickness of the ventricles, detectable ventricular fibrosis may require a prolonged time to develop unless amplified by coexisting pathological conditions, such as hypertension, coronary artery disease, and HF.

IMAGING OF CARDIAC FIBROSIS Significant technological advances allow more possi-

AT1 ¼ angiotensin II receptor type 1; PAI ¼ plasminogen activator inhibitor; other abbreviations as in Tables 1 and 2.

bilities for characterization and quantification of focal and diffuse cardiac fibrosis, which was only possible

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

with biopsy in the past. The most commonly used late

assessed via DE-CMR and a higher CHADS2 (conges-

(delayed) gadolinium enhancement cardiac magnetic

tive heart failure, hypertension, age $75 years, dia-

resonance (DE-CMR) imaging relies on the different

betes mellitus, and stroke/transient ischemic attack)

abilities of healthy myocardium and areas of fibrotic

score and stroke history (89). Moreover, in this pop-

tissue to clear gadolinium, an agent that shortens T1

ulation (n ¼ 387, 36 strokes), adding left atrial

relaxation time. Fibrotic tissue is characterized by

fibrosis to the risk model (i.e., CHADS2 but not ac-

slowing of gadolinium washout, resulting in greater

counting for previous stroke due to the retrospective

signal compared with surrounding “reference” tissue.

nature of the study) improved the C-statistic from

However, finding appropriate reference tissue for

0.58 to 0.72. This outcome was consistent with

quantification of diffuse cardiac fibrosis poses a

another study that demonstrated better prediction of

problem (82,83).

left atrial thrombosis or spontaneous echocardio-

Another cardiac magnetic resonance (CMR)–based

graphic contrast by addition of the degree of atrial

method, T1 mapping, was developed to overcome

fibrosis to either the CHADS2 or CHA2DS2-VASc

this problem. T1 mapping is a calculation of a post-

(congestive heart failure, hypertension, age $75

contrast myocardial T1 time by imaging a given

years, diabetes mellitus, stroke/transient ischemic

plane with sequentially increasing inversion times;

attack, vascular disease, age 65 to 74 years, sex

there is no need to compare the results versus normal

category) scores (90). However, routine use of ex-

reference tissue before or after use of a contrast

pensive, and not universally available, DE-CMR for

agent. This method allows demonstration of diffuse

stroke prediction is not practical at present compared

fibrotic fibers that might appear nearly isointense by

with available and recommended stroke risk ass-

using delayed enhancement (82,83).

essment tools; the approach clearly needs further

Despite being the gold standard, CMR is not widely

validation.

an

Galectin-3 is involved in regulation of fibrosis and

alternative for evaluation of cardiac fibrosis on the

was found to be associated with left atrial volume

basis of the dependency of acoustic properties on

index and AF (odds ratio: 87.5 [95% confidence in-

myocardial composition. Collagen causes ultrasound

terval: 6.1 to 1,265.0]). It was also significantly higher

scattering and attenuation, which can be measured as

in patients with persistent AF than in those with the

integrated backscatter (84). Furthermore, with echo-

paroxysmal type of arrhythmia (91).

available.

Hence,

echocardiography

remains

cardiography, functional assessment as strain peak and strain velocity (parameters that characterize reservoir performance) was found to predict the degree of fibrosis detected in histological specimens and via CMR (85,86).

CLINICAL IMPLICATIONS AND PROGNOSTIC EFFECT OF VENTRICULAR FIBROSIS Ventricular fibrosis has a detrimental effect on both systolic function (due to replacement of apoptotic

CLINICAL IMPLICATIONS AND PROGNOSTIC

and necrotized myocardium) and diastolic function

EFFECTS OF ATRIAL FIBROSIS

(due to increasing stiffness and decreasing compliance) (14). This makes ventricular fibrosis relevant in

Atrial remodeling, including excessive fibrosis, has

the context of HF with both preserved and reduced

major clinical implications in AF. Numerous studies

ejection fraction.

link more extensive atrial interstitial fibrosis to lower

Although scarce data are available on the histo-

effectiveness of AF catheter ablation and the Maze

logical assessment of ventricular myocardial fibrosis

procedure, increased risk of development of post-

in patients with AF, a small case series of ventricular

operative AF, and impaired post-procedural recovery

biopsy specimens suggests the presence of active

of atrial function (Table 4). Interestingly, delayed

myocarditis or nonspecific necrotic/fibrotic changes

enhancement of the atrial myocardium is helpful for

in patients with lone AF (92). It is probable that most

evaluation of post-ablation atrial fibrosis and its

lone AF cases have background myocardial pathology

relation to left atrial reverse remodeling on sinus

that cannot easily be determined with the use of

rhythm (87). Patients undergoing pulmonary vein

routine tests and reflect primary electrical distur-

isolation had a higher AF recurrence rate, with a

bances

lesser degree of left atrial and pulmonary vein scar-

associated with myocardial fibrosis.

ring on DE-CMR (88).

and

subclinical

Contemporary

markers

cardiomyopathies, of

ventricular

likely fibrosis

There are limited data on the association of atrial

(e.g., beta-galactoside–binding lectin galectin-3) are

fibrosis with stroke risk in AF patients. An association

increasingly being studied in patients with HF and

was found between the percentage of atrial fibrosis

have predictive value for adverse outcomes. Their

953

954

T A B L E 4 Studies on the Prognostic Significance of Atrial Fibrosis in AF Patients

Duration of Follow-Up

Evaluation of Atrial Fibrosis

144

283  167 days

DE-MRI

AF catheter ablation

AF recurrence

Intervention

Study Outcome

Association of Atrial Fibrosis and Study Outcome, OR or HR (95% CI)

Increasing recurrence rate with increasing fibrosis degree

Canpolat et al., 2015 (132)

41

18 months

DE-MRI, TGF-b1

AF catheter ablation

AF recurrence

1.127

den Uijl et al., 2011 (133)

170

12  3 months

IBS

AF catheter ablation

AF recurrence

2.80 (2.17–3.61)

Ho et al., 2014 (134)

3,306

10 yrs

Galectin-3

Kornej et al., 2015 (135)

119

6 months

Galectin-3

Kubota et al., 2012 (136)

27

3 yrs

IBS

Kuppahally et al., 2010 (137)

68

Ling et al., 2014 (138)

132

1 yr

Malcolme-Lawes et al., 2013 (139)

50

1 yr

Marrouche et al., 2014 (140)

272

475 days

DE-MRI

McGann et al., 2014 (141)

386

1 yr

DE-MRI

Oakes et al., 2009 (142)

81

9.6  3.7 months

DE-MRI

Olasinska-Wisniewska et al. 2012 (143)

66

12 months

Histology

Park et al., 2013 (144)

128

1 yr

TGF-b1

1, 3, 6, 12, 18, 24, 30 months

NA AF catheter ablation None

Incident AF

1.19 (1.05-1.36)*

AF recurrence

Higher in AF patients but did not predict AF recurrence

Progression from paroxysmal to persistent AF

HR: 8.74 for patients with IBS $20 dB vs. <20 dB

DE-MRI

AF catheter ablation

AF recurrence

1.04 (1.01–1.08)

T1 mapping

AF catheter ablation

AF recurrence

38% of AF recurrence in patients with T1 time <230 ms vs. 25% in patients with T1 time >230 ms

DE-MRI

AF catheter ablation

AF recurrence

Increasing recurrence rate with increasing fibrosis degree

AF catheter ablation

AF recurrence

1.06 (1.03–1.08)

AF catheter ablation

AF recurrence

4.89

AF catheter ablation

AF recurrence

4.88 (1.73–13.74)

AF catheter ablation, mitral valve surgery

AF recurrence

1.09 (1.012–1.17)

Maze procedure

Absence of atrial mechanical contraction

7.47 (1.63–34.4)

Rienstra et al., 2014 (145)

3,217

10 yrs

Soluble ST2

NA

Incident AF

No association

Rosenberg et al., 2014 (146)

2,935

8.8 yrs

PIIINP

NA

Incident AF

0.85 (0.72–1.00) and 0.93 (0.88–0.99) at the 10th and 25th percentiles, setting the median as the reference, no association at the 75th and 90th percentiles

Sasaki et al., 2014 (77)

113

13.8 (8.7–19.9) months

IBS

AF catheter ablation

AF recurrence

1.04 (1.01–1.07)

Seitz et al., 2011 (147)

22

NA

DE-MRI

AF catheter ablation

Wang et al., 2009 (148) Wu et al., 2013 (124)

74

NA

IBS

200

10.9  7.4 months

TGF-b1

CABG AF catheter ablation

“Difficulty” of AF ablation (time to terminate AF; radiofrequency duration until AF termination; complex fractionated atrial electrogram area/LA surface)

Dzeshka et al.

Akoum et al., 2011 (131)

n

Cardiac Fibrosis in Patients With Atrial Fibrillation

First Author, Year (Ref. #)

Significant correlation between the fibrosis grade and the electrophysiological substrate indexes

Post-operative AF

Higher IBS in post-operative AF vs. SR

AF recurrence

1.11 (1.01–1.22)

*Significant association in univariate analysis only, not significant after adjustment for traditional clinical AF risk factors. AUGUST 25, 2015:943–59

JACC VOL. 66, NO. 8, 2015

CABG ¼ coronary artery bypass graft; CI ¼ confidence interval; DE-MRI ¼ delayed enhancement magnetic resonance imaging; HR ¼ hazard ratio; IBS ¼ integrated backscatter; OR ¼ odds ratio; other abbreviations as in Table 1.

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

specific relevance in the context of AF remains to be established (93,94).

Cardiac Fibrosis in Patients With Atrial Fibrillation

Another study that involved patients with arterial hypertension and longer AF duration (median 37

Collagen turnover markers of prognostic signifi-

months) using CMR T1 mapping found the left ven-

cance were assessed in hypertensive heart disease,

tricular extracellular volume to be independently

hypertrophic cardiomyopathy, and HF (95,96). How-

predictive of AF recurrence, as well as of the com-

ever, the major limitation of blood markers of

posite endpoint of AF recurrence admission with HF,

collagen turnover is that they are not cardiac-specific

and death (hazard ratio: 1.35 [95% confidence inter-

and may not accurately reflect myocardial collagen

val: 1.21 to 1.51]) (100). The prognostic value of diffuse

content. Hence, the results are often conflicting,

ventricular fibrosis for AF recurrence after an ablation

depending largely on the selection of study cohorts

procedure was further confirmed by McLellan et al.

(e.g., exclusion of patients with hepatic and kidney

(101). They found that a cutoff level of post-contrast

dysfunction, pulmonary fibrosis, osteoporosis, and

ventricular T1 time <380 ms was associated with

metastatic bone disease, among others, or measuring

better outcome. In a small cohort of patients without

the cardiac gradient of collagen markers [i.e., blood

left ventricular fibrosis, as evidenced by the absence

from coronary sinus vs. systemic circulation]) (97).

of late gadolinium enhancement on CMR in patients

It is even more problematic or even impossible to

with systolic HF, significant improvement of left

distinguish between the atrial and ventricular con-

ventricular function was observed after AF catheter

tributions to circulating levels of byproducts of

ablation. However, the study had no comparator

collagen synthesis and degradation. Consequently,

group with established ventricular fibrosis to assess

attribution of elevated markers of collagen turnover

the effect of sinus rhythm restoration (102).

to AF-related atrial fibrosis alone might not be entirely correct. For example, in the I-PRESERVE (Irbesartan in Heart Failure With Preserved Systolic Function)

Thus far, similar to atrial fibrosis, it is unclear whether AF triggers profibrotic pathways in the left ventricle, is merely a marker of pre-existing fibrotic changes, or both.

collagen substudy that included 29% AF patients, procollagen type I amino-terminal peptide and pro-

TREATMENT APPROACHES TO

collagen type III amino-terminal peptide were pre-

REDUCE CARDIAC FIBROSIS

dictive of all-cause mortality and cardiovascular hospitalizations (98). This association lost significance

Angiotensin II is a potent stimulator of profibrotic

after adjustment for other confounders, however.

pathways, angiotensin-converting enzyme inhibitors,

Similarly, the majority of imaging-based studies on

angiotensin receptor antagonists, and mineralocorti-

the evaluation of ventricular fibrosis focused on pa-

coid receptor antagonists deemed to reduce fibrosis

tients with arterial hypertension, dilated and hyper-

progression. This was supported by several animal

trophic cardiomyopathy, and HF; only a few directly

studies but not, thus far, by clinical trials. Retro-

addressed patients with AF. Moreover, those few

spective analyses and meta-analyses of randomized

studies with AF included mostly patients referred for

trials produced inconclusive results, both for the

AF ablation, which is currently indicated for patients

primary and secondary prevention of AF. It has also

with paroxysmal or persistent arrhythmia resistant to

been accepted that overall primary prevention of AF

antiarrhythmic treatment. Thus, these studies may

with these agents is more feasible than secondary

not be representative of the AF population overall.

prevention (myocardial fibrosis is more likely to be

Recently, Neilan et al. (99) found an association

slowed down than reversed) (103).

between the presence (hazard ratio: 5.08 [95% confi-

Therefore, inhibitors of the angiotensin axis are

dence interval: 3.08 to 8.36]) and extent (hazard

only recommended for AF management when the

ratio: 1.15 [95% confidence interval: 1.10 to 1.21]) of

arrhythmia is associated with other underlying con-

left ventricular late gadolinium enhancement on

ditions that are themselves associated with myocar-

magnetic resonance imaging and all-cause mortality

dial fibrotic remodeling, such as arterial hypertension

(n ¼ 664, median follow-up 42 months, mean left

with left ventricular hypertrophy and systolic HF, and

ventricular ejection fraction 56  10%). The observed

are not recommended in patients with no apparent

associations were even stronger when patients with

cardiovascular disease (e.g., lone AF) (104).

evidence of ischemic heart disease were excluded

Many other components of profibrotic cardiac

from analysis. The major limitation of this study was

pathways (e.g., TGF- b 1, PDGF) represent attractive

the inclusion of patients selected for AF ablation

therapeutic targets. Their suppression with either

with a median duration of arrhythmia since onset of

antibody blockade or oligonucleotide interference

50 days.

reduced interstitial fibrosis in animal experiments.

955

956

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

Nonetheless, experience from the animal data must

atrial and ventricular fibrosis than arrhythmia per se,

be confirmed by clinical trials, and a better under-

but it allows persistent activation of profibrotic

standing of the details of fibrotic pathways is

stimuli. Although the role of atrial fibrosis in AF is

required (15).

well documented, the implication of ventricular fibrosis in pathogenesis and the outcomes of condi-

CONCLUSIONS

tions associated with AF clearly require further

AF is associated with fibrotic processes both in atria

research.

and ventricles. Despite common profibrotic pathways, signaling seems to differ in the ventricular and

REPRINT REQUESTS AND CORRESPONDENCE: Dr.

supraventricular parts of the heart. Atrial fibrosis may

Eduard Shantsila, University of Birmingham Centre

precede development of AF, which in turn results in

for Cardiovascular Sciences, City Hospital, Dudley

further progression of atrial remodeling. Structural

Road, Birmingham, West Midlands B18 7QH, United

heart disease appears to have a greater effect on both

Kingdom. E-mail: [email protected].

REFERENCES 1. Burstein B, Nattel S. Atrial fibrosis: mechanisms and clinical relevance in atrial fibrillation. J Am Coll Cardiol 2008;51:802–9.

14. Khan R, Sheppard R. Fibrosis in heart disease: understanding the role of transforming growth factor-beta in cardiomyopathy, valvular disease

2. Corradi D. Atrial fibrillation from the pathologist’s perspective. Cardiovasc Pathol 2014;23:

and arrhythmia. Immunology 2006;118:10–24. 15. Leask A. Potential therapeutic targets for car-

26. Spinale FG, Coker ML, Thomas CV, et al. Timedependent changes in matrix metalloproteinase

71–84.

diac fibrosis: TGFb, angiotensin, endothelin, CCN2, and PDGF, partners in fibroblast activation. Circ Res 2010;106:1675–80.

activity and expression during the progression of congestive heart failure: relation to ventricular and myocyte function. Circ Res 1998;82:482–95.

16. Greene RM, Nugent P, Mukhopadhyay P, Warner DR, Pisano MM. Intracellular dynamics of Smad-mediated TGFbeta signaling. J Cell Physiol 2003;197:261–71.

27. Kim HE, Dalal SS, Young E, et al. Disruption of the myocardial extracellular matrix leads to cardiac dysfunction. J Clin Invest 2000;106:857–66.

3. McManus DD, Shaikh AY, Abhishek F, et al. Atrial fibrillation and heart failure parallels: lessons for atrial fibrillation prevention. Crit Pathw Cardiol 2011;10:46–51. 4. Oka T, Komuro I. Molecular and cellular mechanisms of organ fibrosis [in Japanese]. Nihon Rinsho 2012;70:1510–6. 5. Tarone G, Balligand JL, Bauersachs J, et al. Targeting myocardial remodelling to develop novel therapies for heart failure: a position paper from the Working Group on Myocardial Function of the European Society of Cardiology. Eur J Heart Fail 2014;16:494–508. 6. Souders CA, Bowers SL, Baudino TA. Cardiac fibroblast: the renaissance cell. Circ Res 2009;105: 1164–76. 7. Zeisberg EM, Kalluri R. Origins of cardiac fibroblasts. Circ Res 2010;107:1304–12. 8. Lajiness JD, Conway SJ. The dynamic role of cardiac fibroblasts in development and disease. J Cardiovasc Transl Res 2012;5:739–48.

17. Grotendorst GR. Connective tissue growth factor: a mediator of TGF-beta action on fibroblasts. Cytokine Growth Factor Rev 1997;8:171–9. 18. Conway SJ, Molkentin JD. Periostin as a heterofunctional regulator of cardiac development and disease. Curr Genomics 2008;9:548–55. 19. Lijnen P, Petrov V. Transforming growth factor-beta1-induced collagen production in cultures of cardiac fibroblasts is the result of the appearance of myofibroblasts. Methods Find Exp Clin Pharmacol 2002;24:333–44. 20. Zhang D, Gaussin V, Taffet GE, et al. TAK1 is activated in the myocardium after pressure overload and is sufficient to provoke heart failure in transgenic mice. Nat Med 2000;6:556–63.

Res 2009;105:934–47.

21. Heger J, Warga B, Meyering B, et al. TGFb receptor activation enhances cardiac apoptosis via SMAD activation and concomitant NO release.

10. Krenning G, Zeisberg EM, Kalluri R. The origin

J Cell Physiol 2011;226:2683–90.

9. Snider P, Standley KN, Wang J, et al. Origin of cardiac fibroblasts and the role of periostin. Circ

of fibroblasts and mechanism of cardiac fibrosis. J Cell Physiol 2010;225:631–7. 11. Andrade J, Khairy P, Dobrev D, et al. The clinical profile and pathophysiology of atrial fibrillation: relationships among clinical features, epidemiology, and mechanisms. Circ Res 2014;114: 1453–68.

22. Rodríguez-Vita J, Sánchez-López E, Esteban V, et al. Angiotensin II activates the Smad pathway in vascular smooth muscle cells by a transforming growth factor-b-independent mechanism. Circulation 2005;111:2509–17.

12. McDowell KS, Vadakkumpadan F, Blake R, et al. Mechanistic inquiry into the role of tissue

23. Mehta PK, Griendling KK. Angiotensin II cell signaling: physiological and pathological effects in the cardiovascular system. Am J Physiol Cell Physiol 2007;292:C82–97.

remodeling in fibrotic lesions in human atrial fibrillation. Biophys J 2013;104:2764–73.

24. Sciarretta S, Paneni F, Palano F, et al. Role of the renin-angiotensin-aldosterone system and in-

13. Baum J, Duffy HS. Fibroblasts and myofibroblasts: what are we talking about? J Cardiovasc Pharmacol 2011;57:376–9.

flammatory processes in the development and progression of diastolic dysfunction. Clin Sci (London) 2009;116:467–77.

25. Kakkar R, Lee RT. Intramyocardial fibroblast myocyte communication. Circ Res 2010;106: 47–57.

28. Spallarossa P, Altieri P, Garibaldi S, et al. Matrix metalloproteinase-2 and -9 are induced differently by doxorubicin in H9c2 cells: the role of MAP kinases and NAD(P)H oxidase. Circ Res 2006;69:736–45. 29. Li YY, McTiernan CF, Feldman AM. Interplay of matrix metalloproteinases, tissue inhibitors of metalloproteinases and their regulators in cardiac matrix remodeling. Cardiovasc Res 2000;46: 214–24. 30. Siddesha JM, Valente AJ, Sakamuri SS, et al. Angiotensin II stimulates cardiac fibroblast migration via the differential regulation of matrixins and RECK. J Mol Cell Cardiol 2013;65: 9–18. 31. Orenes-Piñero E, Montoro-Garcia S, Patel JV, et al. Role of microRNAs in cardiac remodelling: new insights and future perspectives. Int J Cardiol 2013;167:1651–9. 32. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004;116:281–97. 33. Duisters RF, Tijsen AJ, Schroen B, et al. miR133 and miR-30 regulate connective tissue growth factor: implications for a role of microRNAs in myocardial matrix remodeling. Circ Res 2009;104: 170–8, 6p following 178. 34. Liu N, Bezprozvannaya S, Williams AH, et al. microRNA-133a regulates cardiomyocyte proliferation and suppresses smooth muscle gene expression in the heart. Genes Dev 2008;22: 3242–54. 35. Matkovich SJ, Wang W, Tu Y, et al. MicroRNA-133a protects against myocardial fibrosis and modulates electrical repolarization

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

without affecting hypertrophy in pressureoverloaded adult hearts. Circ Res 2010;106: 166–75. 36. Thum T, Gross C, Fiedler J, et al. MicroRNA-21 contributes to myocardial disease by stimulating MAP kinase signalling in fibroblasts. Nature 2008; 456:980–4.

Cardiac Fibrosis in Patients With Atrial Fibrillation

52. Morillo CA, Klein GJ, Jones DL, et al. Chronic rapid atrial pacing. Structural, functional, and electrophysiological characteristics of a new model of sustained atrial fibrillation. Circulation 1995;91:1588–95. 53. He X, Gao X, Peng L, et al. Atrial fibrillation induces myocardial fibrosis through angiotensin II type 1 receptor-specific Arkadia-mediated downregulation of Smad7. Circ Res 2011;108:164–75.

37. Roy S, Khanna S, Hussain SR, et al. MicroRNA expression in response to murine myocardial infarction: miR-21 regulates fibroblast metalloprotease-2 via phosphatase and tensin homologue. Circ Res 2009;82:21–9.

54. Li H, Li S, Yu B, et al. Expression of miR-133 and miR-30 in chronic atrial fibrillation in canines. Mol Med Rep 2012;5:1457–60.

38. Cheng Y, Liu X, Zhang S, et al. MicroRNA-21 protects against the H2O2-induced injury on cardiac myocytes via its target gene PDCD4. J Mol Cell Cardiol 2009;47:5–14.

55. De Jong AM, Van Gelder IC, VreeswijkBaudoin I, et al. Atrial remodeling is directly related to end-diastolic left ventricular pressure in a mouse model of ventricular pressure overload.

39. van Rooij E, Sutherland LB, Thatcher JE, et al.

PloS One 2013;8:e72651.

Dysregulation of microRNAs after myocardial infarction reveals a role of miR-29 in cardiac fibrosis. Proc Natl Acad Sci U S A 2008;105: 13027–32.

56. Everett TH IV, Li H, Mangrum JM, et al. Elec-

40. Hu YF, Chen YJ, Lin YJ, et al. Inflammation and the pathogenesis of atrial fibrillation. Nat Rev Cardiol 2015;12:230–43.

trical, morphological, and ultrastructural remodeling and reverse remodeling in a canine model of chronic atrial fibrillation. Circulation 2000;102: 1454–60.

41. Jalife J. Mechanisms of persistent atrial fibrillation. Curr Opin Cardiol 2014;29:20–7.

57. Li D, Fareh S, Leung TK, et al. Promotion of atrial fibrillation by heart failure in dogs: atrial remodeling of a different sort. Circulation 1999; 100:87–95.

42. Li T, Sun ZL, Xie QY. Meta-analysis identifies serum C-reactive protein as an indicator of atrial

58. Cardin S, Libby E, Pelletier P, et al. Contrasting gene expression profiles in two canine models of

fibrillation risk after coronary artery bypass graft. Am J Ther 2015 Apr 21 [E-pub ahead of print].

atrial fibrillation. Circ Res 2007;100:425–33.

43. Jiang Z, Dai L, Song Z, et al. Association between C-reactive protein and atrial fibrillation recurrence after catheter ablation: a metaanalysis. Clin Cardiol 2013;36:548–54.

59. Rahmutula D, Marcus GM, Wilson EE, et al. Molecular basis of selective atrial fibrosis due to overexpression of transforming growth factor-b1. Cardiovasc Res 2013;99:769–79.

44. 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. J Am

60. Filgueiras-Rama D, Price NF, Martins RP, et al. Long-term frequency gradients during persistent atrial fibrillation in sheep are associated with stable sources in the left atrium. Circ Arrhythm Electrophysiol 2012;5:1160–7.

Coll Cardiol Img 2013;6:672–83.

61. Martins RP, Kaur K, Hwang E, et al. Dominant

45. Li Q, Liu X, Wei J. Ageing related periostin

frequency increase rate predicts transition from paroxysmal to long-term persistent atrial fibrillation. Circulation 2014;129:1472–82.

expression increase from cardiac fibroblasts promotes cardiomyocytes senescent. Biochem Biophys Res Comm 2014;452:497–502. 46. Wu J, Xia S, Kalionis B, et al. The role of oxidative stress and inflammation in cardiovascular aging. BioMed Res Int 2014; 2014:615312.

62. Schoonderwoerd BA, Smit MD, Pen L, et al. New risk factors for atrial fibrillation: causes of ’not-so-lone atrial fibrillation’. Europace 2008;10: 668–73.

47. Boon RA, Iekushi K, Lechner S, et al. MicroRNA-34a regulates cardiac ageing and function. Nature 2013;495:107–10.

63. Grundvold I, Bodegard J, Nilsson PM, et al. Body weight and risk of atrial fibrillation in 7,169 patients with newly diagnosed type 2 dia-

48. Cartledge JE, Kane C, Dias P, et al. Functional crosstalk between cardiac fibroblasts and adult cardiomyocytes by soluble mediators. Cardiovasc Res 2015;105:260–70. 49. Takeda N, Manabe I. Cellular Interplay between cardiomyocytes and nonmyocytes in cardiac remodeling. Int J Inflam 2011;2011:535241. 50. Nattel S, Harada M. Atrial remodeling and atrial fibrillation: recent advances and translational perspectives. J Am Coll Cardiol 2014;63: 2335–45. 51. Sanders P, Morton JB, Davidson NC, et al. Electrical remodeling of the atria in congestive heart failure: electrophysiological and electroanatomic mapping in humans. Circulation 2003; 108:1461–8.

betes; an observational study. Cardiovasc Diabetol 2015;14:5. 64. Pathak RK, Mahajan R, Lau DH, et al. The implications of obesity for cardiac arrhythmia mechanisms and management. Can J Cardiol 2015; 31:203–10. 65. Abed HS, Samuel CS, Lau DH, et al. Obesity results in progressive atrial structural and electrical remodeling: implications for atrial fibrillation. Heart Rhythm 2013;10:90–100. 66. Chilukoti RK, Giese A, Malenke W, et al. Atrial fibrillation and rapid acute pacing regulate

review of the evidence. Cardiol Rev 2015;23: 79–86. 68. Gramley F, Lorenzen J, Jedamzik B, et al. Atrial fibrillation is associated with cardiac hypoxia. Cardiovasc Pathol 2010;19:102–11. 69. Ramos P, Rubies C, Torres M, et al. Atrial fibrosis in a chronic murine model of obstructive sleep apnea: mechanisms and prevention by mesenchymal stem cells. Respir Res 2014;15:54. 70. Iwasaki YK, Kato T, Xiong F, et al. Atrial fibrillation promotion with long-term repetitive obstructive sleep apnea in a rat model. J Am Coll Cardiol 2014;64:2013–23. 71. Neilan TG, Farhad H, Dodson JA, et al. Effect of sleep apnea and continuous positive airway pressure on cardiac structure and recurrence of atrial fibrillation. J Am Heart Assoc 2013;2:e000421. 72. Mahnkopf C, Badger TJ, Burgon NS, et al. Evaluation of the left atrial substrate in patients with lone atrial fibrillation using delayedenhanced MRI: implications for disease progression and response to catheter ablation. Heart Rhythm 2010;7:1475–81. 73. Olesen MS, Nielsen MW, Haunsø S, et al. Atrial fibrillation: the role of common and rare genetic variants. Eur J Hum Genet 2014;22:297–306. 74. Anumonwo JM, Kalifa J. Risk factors and genetics of atrial fibrillation. Cardiol Clin 2014;32: 485–94. 75. Chinchilla A, Daimi H, Lozano-Velasco E, et al. PITX2 insufficiency leads to atrial electrical and structural remodeling linked to arrhythmogenesis. Circ Cardiovasc Genet 2011;4:269–79. 76. Ling LH, Kistler PM, Ellims AH, et al. Diffuse ventricular fibrosis in atrial fibrillation: noninvasive evaluation and relationships with aging and systolic dysfunction. J Am Coll Cardiol 2012;60: 2402–8. 77. Sasaki N, Okumura Y, Watanabe I, et al. Transthoracic echocardiographic backscatterbased assessment of left atrial remodeling involving left atrial and ventricular fibrosis in patients with atrial fibrillation. Int J Cardiol 2014; 176:1064–6. 78. Shantsila E, Shantsila A, Blann AD, et al. Left ventricular fibrosis in atrial fibrillation. Am J Cardiol 2013;111:996–1001. 79. Chrysostomakis SI, Karalis IK, Simantirakis EN, et al. Angiotensin II type 1 receptor inhibition is associated with reduced tachyarrhythmia-induced ventricular interstitial fibrosis in a goat atrial fibrillation model. Cardiovasc Drugs Ther 2007;21: 357–65. 80. Avitall B, Bi J, Mykytsey A, Chicos A. Atrial and ventricular fibrosis induced by atrial fibrillation: evidence to support early rhythm control. Heart Rhythm 2008;5:839–45. 81. Burstein B, Libby E, Calderone A, et al. Differential behaviors of atrial versus ventricular fibroblasts: a potential role for platelet-derived growth factor in atrial-ventricular remodeling

adipocyte/adipositas-related gene expression in the atria. Int J Cardiol 2015;187:604–13.

differences. Circulation 2008;117:1630–41.

67. Matassini MV, Brambatti M, Guerra F, et al. Sleep-disordered breathing and atrial fibrillation:

central prognostic tool in myocardial fibrosis. Nat Rev Cardiol 2015;12:18–29.

82. Ambale-Venkatesh B, Lima JA, Cardiac MRI. a

957

958

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

Cardiac Fibrosis in Patients With Atrial Fibrillation

83. Iles LM, Ellims AH, Llewellyn H, et al. Histological validation of cardiac magnetic resonance analysis of regional and diffuse interstitial myocardial fibrosis. Eur Heart J Cardiovasc Imaging 2015;16:14–22. 84. Zhu H, Zhang W, Zhong M, et al. Myocardial ultrasonic integrated backscatter analysis in patients with chronic atrial fibrillation. Int J Cardiovasc Imaging 2010;26:861–5. 85. Longobardo L, Todaro MC, Zito C, et al. Role of imaging in assessment of atrial fibrosis in patients with atrial fibrillation: state-of-the-art review. Eur Heart J Cardiovasc Imaging 2014;15:1–5. 86. Habibi M, Lima JA, Khurram IM, et al. Association of left atrial function and left atrial enhancement in patients with atrial fibrillation: cardiac magnetic resonance study. Circ Cardiovasc Imaging 2015;8:e002769. 87. Bax JJ, Marsan NA, Delgado V. Non-invasive imaging in atrial fibrillation: focus on prognosis and catheter ablation. Heart 2015;101:94–100. 88. Peters DC, Wylie JV, Hauser TH, et al. Recurrence of atrial fibrillation correlates with the extent of post-procedural late gadolinium enhancement: a pilot study. J Am Coll Cardiol Img 2009;2:308–16. 89. Daccarett M, Badger TJ, Akoum N, et al. Association of left atrial fibrosis detected by delayed-enhancement magnetic resonance imaging and the risk of stroke in patients with atrial fibrillation. J Am Coll Cardiol 2011;57:831–8. 90. Akoum N, Fernandez G, Wilson B, et al. Association of atrial fibrosis quantified using LGE-MRI with atrial appendage thrombus and spontaneous contrast on transesophageal echocardiography in patients with atrial fibrillation. J Cardiovasc Electrophysiol 2013;24:1104–9. 91. Gurses KM, Yalcin MU, Kocyigit D, et al. Effects of persistent atrial fibrillation on serum galectin-3 levels. Am J Cardiol 2015;115:647–51. 92. Frustaci A, Caldarulo M, Buffon A, et al. Cardiac biopsy in patients with “primary” atrial fibrillation: histologic evidence of occult myocardial diseases. Chest 1991;100:303–6. 93. Bayes-Genis A, de Antonio M, Vila J, et al. Head-to-head comparison of 2 myocardial fibrosis biomarkers for long-term heart failure risk stratification: ST2 versus galectin-3. J Am Coll Cardiol 2014;63:158–66. 94. Lopez-Andrès N, Rossignol P, Iraqi W, et al. Association of galectin-3 and fibrosis markers with long-term cardiovascular outcomes in patients with heart failure, left ventricular dysfunction, and dyssynchrony: insights from the CARE-HF (Cardiac Resynchronization in Heart Failure) trial. Eur J Heart Fail 2012;14:74–81. 95. Querejeta R, López B, González A, et al. Increased collagen type I synthesis in patients with heart failure of hypertensive origin: relation to myocardial fibrosis. Circulation 2004;110:1263–8.

myocardial matrix metabolism in advanced heart failure. Eur J Heart Fail 2013;15:292–8.

of the TGF-b1 signaling pathway. Int J Cardiol 2010;143:405–13.

98. Krum H, Elsik M, Schneider HG, et al. Relation of peripheral collagen markers to death and hospitalization in patients with heart failure and preserved ejection fraction: results of the I-PRESERVE collagen substudy. Circ Heart Fail 2011;4:561–8.

111. Kallergis EM, Manios EG, Kanoupakis EM, et al. Extracellular matrix alterations in patients

99. Neilan TG, Shah RV, Abbasi SA, et al. The incidence, pattern, and prognostic value of left ventricular myocardial scar by late gadolinium enhancement in patients with atrial fibrillation.

112. Ko WC, Hong CY, Hou SM, et al. Elevated expression of connective tissue growth factor in human atrial fibrillation and angiotensin II-treated

J Am Coll Cardiol 2013;62:2205–14. 100. Neilan TG, Mongeon FP, Shah RV, et al. Myocardial extracellular volume expansion and the risk of recurrent atrial fibrillation after pulmonary vein isolation. J Am Coll Cardiol Img 2014; 7:1–11. 101. McLellan AJ, Ling LH, Azzopardi S, et al. Diffuse ventricular fibrosis measured by T1 mapping on cardiac MRI predicts success of catheter ablation for atrial fibrillation. Circ Arrhythm Electrophysiol 2014;7:834–40. 102. Ling LH, Taylor AJ, Ellims AH, et al. Sinus rhythm restores ventricular function in patients with cardiomyopathy and no late gadolinium enhancement on cardiac magnetic resonance im-

with paroxysmal and persistent atrial fibrillation: biochemical assessment of collagen type-I turnover. J Am Coll Cardiol 2008;52:211–5.

cardiomyocytes. Circ J 2011;75:1592–600. 113. Li Y, Jian Z, Yang ZY, et al. Increased expression of connective tissue growth factor and transforming growth factor-beta-1 in atrial myocardium of patients with chronic atrial fibrillation. Cardiology 2013;124:233–40. 114. Mayyas F, Niebauer M, Zurick A, et al. Association of left atrial endothelin-1 with atrial rhythm, size, and fibrosis in patients with structural heart disease. Circ Arrhythm Electrophysiol 2010;3:369–79. 115. Nishi H, Sakaguchi T, Miyagawa S, et al. Impact of microRNA expression in human atrial tissue in patients with atrial fibrillation undergoing cardiac surgery. PloS One 2013;8:e73397.

aging who undergo catheter ablation for atrial fibrillation. Heart Rhythm 2013;10:1334–9.

116. Okumura Y, Watanabe I, Nakai T, et al. Impact of biomarkers of inflammation and extracellular matrix turnover on the outcome of atrial fibrillation

103. Savelieva I, Kakouros N, Kourliouros A, et al. Upstream therapies for management of atrial

ablation: importance of matrix metalloproteinase-2 as a predictor of atrial fibrillation recurrence. J Cardiovasc Electrophysiol 2011;22:987–93.

fibrillation: review of clinical evidence and implications for European Society of Cardiology guidelines. Part I: primary prevention. Europace 2011;13:308–28. 104. Camm AJ, Kirchhof P, Lip GY, et al., for the European Heart Rhythm Association; European Association for Cardio-Thoracic Surgery. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010;31:2369–429. 105. Adam O, Lavall D, Theobald K, et al. Rac1induced connective tissue growth factor regulates connexin 43 and N-cadherin expression in atrial fibrillation. J Am Coll Cardiol 2010;55: 469–80. 106. Adam O, Löhfelm B, Thum T, et al. Role of miR-21 in the pathogenesis of atrial fibrosis. Basic Res Cardiol 2012;107:278. 107. Cao H, Li Q, Li M, et al. Osteoprotegerin/ RANK/RANKL axis and atrial remodeling in mitral valvular patients with atrial fibrillation. Int J Cardiol 2013;166:702–8. 108. Dawson K, Wakili R, Ordög B, et al. MicroRNA29: a mechanistic contributor and potential biomarker in atrial fibrillation. Circulation 2013; 127:1466–75, 1475e1–28. 109. Gramley F, Lorenzen J, Plisiene J, et al. Decreased plasminogen activator inhibitor and tissue metalloproteinase inhibitor expression may promote increased metalloproteinase activity with increasing duration of human atrial fibrillation.

96. Kosmala W, Przewlocka-Kosmala M, Szczepanik-Osadnik H, et al. Fibrosis and cardiac function in obesity: a randomised controlled trial of aldosterone blockade. Heart 2013;99:320–6.

J Cardiovasc Electrophysiol 2007;18:1076–82.

97. Kaye DM, Khammy O, Mariani J, et al. Relationship of circulating matrix biomarkers to

110. Gramley F, Lorenzen J, Koellensperger E, et al. Atrial fibrosis and atrial fibrillation: the role

117. Polyakova V, Miyagawa S, Szalay Z, et al. Atrial extracellular matrix remodelling in patients with atrial fibrillation. J Cell Mol Med 2008;12: 189–208. 118. Qu YC, Du YM, Wu SL, et al. Activated nuclear factor-kB and increased tumor necrosis factor-a in atrial tissue of atrial fibrillation. Scand Cardiovasc J 2009;43:292–7. 119. Richter B, Gwechenberger M, Socas A, et al. Time course of markers of tissue repair after ablation of atrial fibrillation and their relation to left atrial structural changes and clinical ablation outcome. Int J Cardiol 2011;152:231–6. 120. Rudolph V, Andrié RP, Rudolph TK, et al. Myeloperoxidase acts as a profibrotic mediator of atrial fibrillation. Nat Med 2010;16:470–4. 121. Swartz MF, Fink GW, Sarwar MF, et al. Elevated pre-operative serum peptides for collagen I and III synthesis result in post-surgical atrial fibrillation. J Am Coll Cardiol 2012;60: 1799–806. 122. Wang J, Wang Y, Han J, et al. Integrated analysis of microRNA and mRNA expression profiles in the left atrium of patients with nonvalvular paroxysmal atrial fibrillation: role of miR-146b-5p in atrial fibrosis. Heart Rhythm 2015;12:1018–26. 123. Wilhelm M, Kirste W, Kuly S, et al. Atrial distribution of connexin 40 and 43 in patients with intermittent, persistent, and postoperative atrial fibrillation. Heart Lung Circ 2006;15:30–7. 124. Wu CH, Hu YF, Chou CY, et al. Transforming growth factor-b1 level and outcome after catheter ablation for nonparoxysmal atrial fibrillation. Heart Rhythm 2013;10:10–5.

Dzeshka et al.

JACC VOL. 66, NO. 8, 2015 AUGUST 25, 2015:943–59

125. Xi L, Cao H, Zhu J, et al. OPG/RANK/RANKL axis in stabilization of spontaneously restored sinus rhythm in permanent atrial fibrillation patients after mitral valve surgery. Cardiology 2013;124: 18–24. 126. Xie X, Liu Y, Gao S, et al. Possible involvement of fibrocytes in atrial fibrosis in patients with chronic atrial fibrillation. Circ J 2014;78:338–44. 127. Xu GJ, Gan TY, Tang BP, et al. Differential expression of collagen and matrix metalloproteinases between left and right atria in patients with chronic atrial fibrillation [in Chinese]. Sheng Li Xue Bao 2009;61:211–6. 128. Cardin S, Guasch E, Luo X, et al. Role for MicroRNA-21 in atrial profibrillatory fibrotic remodeling associated with experimental postinfarction heart failure. Circ Arrhythm Electrophysiol 2012;5:1027–35. 129. Saba S, Janczewski AM, Baker LC, et al. Atrial contractile dysfunction, fibrosis, and arrhythmias in a mouse model of cardiomyopathy secondary to cardiac-specific overexpression of tumor necrosis factor-a. Am J Physiol Heart Circ Physiol 2005; 289:H1456–67.

Cardiac Fibrosis in Patients With Atrial Fibrillation

133. den Uijl DW, Delgado V, Bertini M, et al. Impact of left atrial fibrosis and left atrial size on the outcome of catheter ablation for atrial fibrillation. Heart 2011;97:1847–51. 134. Ho JE, Yin X, Levy D, et al. Galectin 3 and incident atrial fibrillation in the community. Am Heart J 2014;167:729–34.e1. 135. Kornej J, Schmidl J, Ueberham L, et al. Galectin-3 in patients with atrial fibrillation undergoing radiofrequency catheter ablation. PloS One 2015;10:e0123574. 136. Kubota T, Kawasaki M, Takasugi N, et al. Left atrial pathological degeneration assessed by integrated backscatter transesophageal echocardiography as a predictor of progression to persistent atrial fibrillation: results from a prospective study of three-years follow-up. Cardiovasc Ultrasound 2012;10:28. 137. Kuppahally SS, Akoum N, Badger TJ, et al. Echocardiographic left atrial reverse remodeling after catheter ablation of atrial fibrillation is predicted by preablation delayed enhancement of left atrium by magnetic resonance imaging. Am Heart J 2010;160:877–84.

130. Verheule S, Sato T, Everett TIV, et al. Increased vulnerability to atrial fibrillation in transgenic mice with selective atrial fibrosis caused by overexpression of TGF-b1. Circ Res 2004;94:1458–65.

138. Ling LH, McLellan AJ, Taylor AJ, et al. Magnetic resonance post-contrast T1 mapping in the human atrium: validation and impact on clinical outcome after catheter ablation for atrial fibrilla-

131. Akoum N, Daccarett M, McGann C, et al. Atrial fibrosis helps select the appropriate patient and strategy in catheter ablation of atrial fibrillation: a DE-MRI guided approach. J Cardiovasc Electrophysiol 2011;22:16–22.

139. Malcolme-Lawes LC, Juli C, Karim R, et al. Automated analysis of atrial late gadolinium enhancement imaging that correlates with endocardial voltage and clinical outcomes: a 2-center study. Heart Rhythm 2013;10:1184–91.

132. Canpolat U, Oto A, Hazirolan T, et al. A prospective DE-MRI study evaluating the role of TGF-b1 in left atrial fibrosis and implications for outcomes of cryoballoon-based catheter ablation: new insights into primary fibrotic atriocardiomyopathy. J Cardiovasc Electrophysiol 2015;26: 251–9.

tion. Heart Rhythm 2014;11:1551–9.

140. Marrouche NF, Wilber D, Hindricks G, et al. Association of atrial tissue fibrosis identified by delayed enhancement MRI and atrial fibrillation catheter ablation: the DECAAF study. JAMA 2014; 311:498–506. 141. McGann C, Akoum N, Patel A, et al. Atrial fibrillation ablation outcome is predicted by left

atrial remodeling on MRI. Circ Arrhythm Electrophysiol 2014;7:23–30. 142. Oakes RS, Badger TJ, Kholmovski EG, et al. Detection and quantification of left atrial structural remodeling with delayed-enhancement magnetic resonance imaging in patients with atrial fibrillation. Circulation 2009;119:1758–67. 143. Olasinska-Wisniewska A, Mularek-Kubzdela T, Grajek S, et al. Impact of atrial remodeling on heart rhythm after radiofrequency ablation and mitral valve operations. Ann Thorac Surg 2012;93: 1449–55. 144. Park SJ, On YK, Kim JS, et al. Transforming growth factor b1-mediated atrial fibrotic activity and the recovery of atrial mechanical contraction after surgical maze procedure. Int J Cardiol 2013; 164:232–7. 145. Rienstra M, Yin X, Larson MG, et al. Relation between soluble ST2, growth differentiation factor-15, and high-sensitivity troponin I and incident atrial fibrillation. Am Heart J 2014;167: 109–15.e2. 146. Rosenberg MA, Maziarz M, Tan AY, et al. Circulating fibrosis biomarkers and risk of atrial fibrillation: the Cardiovascular Health Study (CHS). Am Heart J 2014;167:723–8.e2. 147. Seitz J, Horvilleur J, Lacotte J, et al. Correlation between AF substrate ablation difficulty and left atrial fibrosis quantified by delayed-enhancement cardiac magnetic resonance. Pacing Clin Electrophysiol 2011;34:1267–77. 148. Wang GD, Shen LH, Wang L, et al. Relationship between integrated backscatter and atrial fibrosis in patients with and without atrial fibrillation who are undergoing coronary bypass surgery. Clin Cardiol 2009;32: E56–61.

KEY WORDS cardiac, extracellular matrix, heart failure, myocytes, myofibroblasts

959