+Model

ACVD-806; No. of Pages 8 ARTICLE IN PRESS

Archives of Cardiovascular Disease (2015) xxx, xxx—xxx

Available online at

ScienceDirect
www.sciencedirect.com

REVIEW

Type 2 diabetes mellitus and atrial

fibrillation: From mechanisms to clinical
practice
Diabète de type II et fibrillation atriale : des mécanismes à la pratique clinique

Marijana Tadic a,b,∗, Cesare Cuspidi c

a
University Clinical Hospital Centre ‘‘Dr. Dragisa Misovic - Dedinje’’, Belgrade, Serbia
b
Faculty of Medicine, University of Belgrade, Belgrade, Serbia
c
Clinical Research Unit, University of Milan-Bicocca and Istituto Auxologico Italiano,
Meda, Italy

Received 11 November 2014; received in revised form 26 January 2015; accepted 26 January
2015

KEYWORDS Summary Type 2 diabetes mellitus is one of the most common chronic conditions and its
Type 2 diabetes prevalence has increased continuously over the past decades, primarily due to the obesity
mellitus; epidemic. Atrial fibrillation (AF) is the most frequent sustained cardiac arrhythmia in clinical
Atrial fibrillation; practice and is associated with increased cardiovascular and cerebrovascular morbidity and
Mechanisms; mortality. Recent studies have shown that patients with diabetes have an increased risk of
Treatment; AF. However, the results about the relationship between diabetes and AF are still conflicting.
Clinical studies Mechanisms that are responsible for an association between diabetes and AF, as well as the
adequate treatment of AF in patients with diabetes, are still insufficiently studied. The aim
of this review is to summarize the current knowledge of mechanisms that connect AF and
diabetes, the clinical studies that include patients with both conditions, and the treatment
options in modern pharmacology.
© 2015 Elsevier Masson SAS. All rights reserved.

Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; AF, atrial fibrillation; ARB, angiotensin II receptor blocker; CI, confidence
interval; HOMA, homeostatic model assessment; RAAS, renin-angiotensin-aldosterone system.
∗ Corresponding author at: University Clinical Hospital Center ‘‘Dr. Dragisa Misovic - Dedinje’’, Heroja Milana Tepica 1, 11000 Belgrade,

Serbia.
E-mail address: marijana tadic@hotmail.com (M. Tadic).

http://dx.doi.org/10.1016/j.acvd.2015.01.009
1875-2136/© 2015 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
2 M. Tadic, C. Cuspidi

MOTS CLÉS Résumé Le diabète de type II est une des affections chroniques les plus fréquentes et sa pré-
Fibrillation atriale ; valence a augmenté de façon régulière au cours des dernières décennies, essentiellement du
Mécanismes ; fait de l’épidémie d’obésité. La fibrillation atriale est l’arythmie cardiaque la plus fréquente en
Traitement ; pratique clinique et est associée avec une augmentation de la morbi-mortalité cardiovasculaire
Études cliniques et cérébrovasculaire. Des études récentes ont montré que les patients diabétiques avaient
un risque accru de fibrillation atriale. Cependant, ces résultats établissant la relation entre
diabète et fibrillation atriale sont controversés. Les mécanismes responsables d’une telle asso-
ciation, ainsi que le traitement de la fibrillation atriale chez les diabétiques sont insuffisamment
étudiés à ce jour. L’objectif de cette revue générale est de résumer les données actuelles des
mécanismes liant la fibrillation atriale au diabète, ainsi que les études cliniques qui incluent
les patients ayant un diabète et une fibrillation atriale ; les options thérapeutiques et l’apport
de la pharmacologie moderne sont également discutés.
© 2015 Elsevier Masson SAS. Tous droits réservés.

Introduction [2,7—9]. Diabetes could also cause structural, electrical,

electromechanical and autonomic remodelling [10].
Type 2 diabetes mellitus is one of the most common chronic
conditions and its prevalence is increasing. In 2014, the num- Atrial structural remodelling
ber of patients with diabetes was estimated at 387 millions
people worldwide, of whom around half were undiagnosed Most evidence obtained from animal studies has demon-
[1]. The projection for 2035 is that the number of patients strated that structural remodelling of the left atrium,
with diabetes will have risen to 592 millions [1]. It is well primarily atrial dilatation and interstitial fibrosis is the
known that cardiovascular and cerebrovascular diseases are major trigger of AF in patients with diabetes [11,12]. Struc-
substantially more frequent among patients with diabetes tural remodelling leads to ionic remodelling and increases
[2]. However, the mechanisms of this influence are still not atrial tissue vulnerability, which induces inter-atrial conduc-
completely understood. It is not known whether permanent tion delay and, subsequently, initiation and maintenance of
hyperglycaemia or wide fluctuations in glucose levels—from AF [11,12].
hypoglycaemia to hyperglycaemia, which is commonly seen Atrial fibrosis could be caused by inflammation (oxida-
in diabetes subjects, especially the treated ones—is most tive stress), increased production of advanced glycation
responsible for cardiovascular damage. end products, increased expression of transforming growth
Atrial fibrillation (AF), the most frequent sustained car- factor-␤ and expression changes of gap junction proteins
diac arrhythmia, is associated with an increased risk of [10].
stroke, thromboembolism, heart failure and recurrent hos-
pital admissions [3]. The prevalence of AF is constantly
increasing, and it affects about 2.2 million individuals in Atrial electrical remodelling
America and 4.5 millions in Europe [4,5].
The principal characters of atrial electrical remodelling
Studies have shown that diabetes is frequently associated
involve atrial effective refractory period shortening and its
with AF [6]. However, it is difficult to determine whether
dispersion, as well as impairment of its frequency adaptation
diabetes directly affects the atrial tissue or whether differ-
and consequent inter-atrial conduction delay [10]. Changes
ent pathways are involved, including hypertension, coronary
in left atrial diameter, atrial activation time and voltage
artery disease and abnormal activity of the autonomic ner-
in patients with diabetes have been associated with atrial
vous system. The aim of this review is to summarize current
structural remodelling, not only electrical, especially in
knowledge of the relationships between diabetes and AF,
patients with AF. Chao et al., for example, studied patients
from mechanisms to clinically relevant studies.
with abnormal glucose metabolism (diabetes and impaired
glucose tolerance) who underwent radiofrequency ablation
of paroxysmal AF for the first time, and revealed that right
Mechanisms of the relationship between and left atria voltages in these patients were significantly
diabetes and AF decreased due to atrial electrical remodelling and atrial
fibrosis [13].
Diabetes is associated with numerous metabolic defects
including insulin resistance, impaired glucose tolerance, Atrial electromechanical remodelling
proinflammatory mediators, abnormalities of haemostasis,
fibrinolysis, angiogenesis and extracellular matrix turnover Findings from our group show that left and right atrial reser-
[7—9] (Fig. 1). All of these metabolic changes lead to voir and conduit function, assessed by emptying fractions
endothelial dysfunction, abnormal activation of the renin- and strain, are decreased among subjects with prediabetes
angiotensin-aldosterone system (RAAS) and acceleration of and patients with diabetes compared with controls [14]. On
atherogenesis, which could be responsible for AF occurrence the other hand, left and right atrial pump function increased

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
Diabetes and atrial fibrillation 3

Figure 1. Possible mechanisms that link diabetes and atrial fibrillation occurrence. RAAS: renin-angiotensin-aldosterone system.

in the same direction. These results agree with previous Benjamin et al. [3] and Nichols et al. [27] revealed that DM
studies [15,16]. Furthermore, Ayhan et al. found similar increases the risk of AF to a greater extent in women than in
results regarding left atrial function, and also revealed that men [3,27]. A recently published study showed that a signif-
inter- and intra-atrial electromechanical delays were signif- icant relationship between baseline diabetes and incident
icantly higher in patients with glucose intolerance than in AF was the consequence of changes of other AF risk factors
controls [17]. Studies have also shown that that inter- and (age, hypertension, obesity) [28].
intra-atrial electromechanical abnormalities are associated In a Swedish community-based study, the authors demon-
with fasting glucose level and AF occurrence in patients with strated that patients with arterial hypertension and diabetes
diabetes [11,12]. had a threefold higher risk of AF occurrence compared with
normotensive and non-diabetic control subjects [29]. How-
Atrial autonomic remodelling ever, after adjustment for homeostatic model assessment
(HOMA) index, the relationship was no longer statistically
The influence of diabetes on the autonomic nervous significant, which implies that insulin resistance could be
system—sympathetic and parasympathetic—is well known the main underlying mechanism linking diabetes, hyperten-
[18,19]. Studies have demonstrated that atrial tissue in sion and AF [29]. The VALUE trial also demonstrated that
patients with diabetes has a greater ability to uptake choline hypertension and diabetes have a synergistic effect on AF
and release acetylcholine [20]. development [30].
Otake et al. revealed that diabetic mice were more sus- Movahed et al. showed that diabetes was independently
ceptible to AF development after sympathetic stimulation associated with AF and atrial flutter, as well as left ventricu-
than were controls [21]. The electrophysiological investi- lar hypertrophy, coronary artery disease and chronic heart
gation demonstrated shortened atrial effective refractory failure [6]. Dublin et al. reported that risk of AF develop-
period and increased dispersion. However, parasympathetic ment was higher with longer duration of diabetes and worse
stimulation in diabetic mice did not affect the atrial glycaemic control [23].
effective refractory period or the occurrence of AF [21]. The ARIC study showed that elevated blood pressure
These findings imply that neural remodelling could have an was the most important contributor to the burden of AF,
essential role in the development of AF in patients with while merely 3% of AF risk was explained by diabetes [31].
diabetes. Thacker et al. found that diabetes, hypertension and blood
pressure were not associated with permanent AF [32]. The
CARAF study also showed that diabetes did not contribute
Clinical studies about diabetes and AF to the progression from paroxysmal to chronic AF [33]. Sim-
ilar results were obtained by Tsang et al. in the Olmstead
Clinical investigations do not completely agree about the County study [34]. Interestingly, the NAVIGATOR investiga-
influence of diabetes on AF occurrence. Actually, the find- tors showed that impaired glucose tolerance and fasting
ings of these studies are conflicting (Table 1). Some authors plasma glucose, but not progression to diabetes, predict the
found that diabetes increased the risk of AF occurrence risk of AF [35]. On the other hand, the results of the Framing-
[6,22,23], whereas other studies did not reveal statistically ham study demonstrated no significant association between
significant differences in the risk of AF development with insulin resistance and incident AF [36].
diabetes [24,25]. Interestingly, some authors investigated A large meta-analysis has shown that the patients with
differences in AF occurrence between men and women with diabetes have a nearly 40% greater risk of AF comparing
diabetes. Frost et al. did not find any difference in AF preva- to non-diabetic patients (relative risk 1.39, 95% confidence
lence between men and women with diabetes [26], whereas interval [CI] 1.10—1.75; P < 0.001) [37].

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
4 M. Tadic, C. Cuspidi

Table 1 The effect of diabetes on AF occurrence.

Reference Sample size and subjects included Main findings
in the study
Movahed et al. 293,124 diabetic patients and Diabetes was a strong, independent
[6] 552,624 hypertensive patients risk factor for the occurrence of AF
and atrial flutter, heart failure, left
ventricular hypertrophy and CAD
Manitoba 3983 male air crew recruits Diabetes was a significant risk factor
Follow-Up observed continuously for 44 years; for AF development
Study [22] 299 developed AF
Dublin et al. [23] 1410 people with newly recognized Diabetes was associated with AF
AF occurrence; risk was higher with
longer duration of treated diabetes
and worse glycaemic control
Ruigomez et al. 1035 confirmed incident cases of Diabetes was not a risk factor of AF
[24] chronic AF and 5000 controls from occurrence
the general population
Danish Diet, 47,589 participants without Intermediate occurrence of diabetes
Cancer, and pre-existing cardiovascular or during follow-up did not increase the
Health Study endocrine disease risk for AF development
[26]
Framingham 2090 men and 2641 women, no Men had a 1.5 times greater risk of
Heart Study [3] history of AF developing AF than women; diabetes
was a more important risk factor for
AF in women
Nichols et al. [27] 17,372 patients with diabetes and Diabetes was an independent
matched subjects without diabetes predictor of AF among women but
not men
Schoen et al. [28] 34,720 female health professionals A significant relationship between
who were free of cardiovascular baseline diabetes and incident AF
disease and AF at baseline could be mainly explained by other
cardiovascular risk factors
Swedish 1739 subjects of both genders Diabetes was not associated with AF,
community whereas the combination of
study [29] hypertension and diabetes was
related to AF
ARIC study [31] 14,598 middle-aged people Diabetes could explain 3% of the
followed for 17.1 years overall risk for AF occurrence
Thacker et al. 1385 patients with newly diagnosed Diabetes, blood pressure and
[32] AF whose initial AF terminated hypertension were not associated
within 6 months and who with permanent AF
had ≥ 6 months of subsequent
follow-up
CARAF [33] 757 patients with a baseline Diabetes did not contribute to the
diagnosis of paroxysmal AF progression from paroxysmal to
chronic AF
Tsang et al. [34] 3248 with paroxysmal AF Progression from paroxysmal to
permanent AF was related to gradual
BMI increase, but not diabetes
NAVIGATOR study 8943 patients without AF at Impaired glucose tolerance, but not
[35] baseline progression to diabetes, predicted
risk of AF
Huxley et al. [37] Meta-analysis including 108,703 Patients with diabetes had an
cases of AF in 1,686,097 subjects approximate 40% greater risk of AF
compared to unaffected patients;
the population-attributable fraction
of AF due to diabetes was 2.5%
AF: atrial fibrillation; CAD: coronary artery disease.

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
Diabetes and atrial fibrillation 5

Table 2 The impact of diabetes on AF recurrence.

Reference Sample size and subjects included in the Main findings
study
Forleo et al. [38] 70 patients with diabetes and paroxysmal Catheter ablation of AF provides
or persistent AF treated with catheter significant clinical benefits over
ablation antiarrhythmic drugs in diabetic patients
Chao et al. [13] 228 patients with paroxysmal AF who had Abnormal glucose metabolism induced an
undergone catheter ablation intra-atrial conduction delay, reduced
voltage and greater recurrence rate after
catheter ablation
Chang et al. [39] 282 patients with AF who underwent Metabolic syndrome was associated with
catheter ablation AF recurrence after AF ablation
Cai et al. [40] 161 paroxysmal AF and 25 Diabetes correlated with AF recurrence,
non-paroxysmal AF patients who but not independently of other
underwent catheter ablation cardiovascular risk factors
D’Ascenzo et al. [41] Meta-analysis including 4357 patients Diabetes was an independent predictor
with paroxysmal AF, 1083 with persistent of AF recurrence
AF and 1777 with long-standing AF
AF: atrial fibrillation.

The influence of diabetes on recurrent AF Diabetes, AF and modern management

after catheter ablation
The management of AF involves control of symptoms and
Catheter ablation, that includes the isolation of the prevention of complications. In patients with diabetes, it is
pulmonary vein, has been established as treatment for even more difficult to achieve these goals.
drug-resistant paroxysmal AF. However, recurrence of AF In the Framingham study, AF was proven to be a
is common, and repeated ablation procedures are still prothrombotic state [3], thus anticoagulant therapy is con-
required in a significant percentage of patients for recurrent sidered crucial for the prevention of stroke; and guidelines
AF. The role of diabetes in AF recurrence is still unknown, recommend it for all AF patients with at least one risk factor
and the findings are conflicting (Table 2). for stroke [42]. Considering the fact that diabetes is a pro-
Forleo et al. reported that catheter ablation of AF in thrombotic state and one of the risk factors of stroke, one
patients with diabetes provided significant clinical bene- could conclude that all patients with these two comorbidi-
fits over the antiarrhythmic drugs [38]. They concluded that ties (diabetes and AF) should be treated with anticoagulants.
catheter ablation was a reasonable approach regarding fea- Hart et al. demonstrated that patients with paroxysmal, per-
sibility, effectiveness and low procedural risk. sistent or permanent AF had the same risk of stroke [43],
Chao et al. demonstrated that AF recurrence rate which should especially be kept in mind in patients with
was greater among patients with an abnormal glucose diabetes.
metabolism than in those without it (18.5% vs 8.0%; Warfarin has long been the anticoagulant of choice in
P = 0.022) [13]. The authors explained these findings by patients with AF. However, a low percentage of patients who
increased left atrial size in diabetes subjects, which could reach treatment goals (< 40%), need close monitoring, risk
contribute to the heterogeneity of the left atrium and initi- of treatment-related bleeding and interference with other
ation of AF. drugs and food have made the use of warfarin challenging.
Investigations have revealed that patients with metabolic On the other hand, new anticoagulants such as direct throm-
syndrome (with insulin resistance as a main feature) had a bin inhibitors (dabigatran) and direct factor Xa inhibitors
higher incidence of recurrent AF compared with those with- (apixaban, rivaroxaban, edoxaban and betrixaban), have
out metabolic syndrome [39]. Diabetes was present in 28% of appeared on the market with much better pharmacokinetic
the subjects with metabolic syndrome. A higher incidence of profiles.
recurrent AF from non-pulmonary vein origin was observed Dabigatran—a direct thrombin inhibitor—has a good
among patients with metabolic syndrome compared with safety profile, a low potential for drug and food interac-
those without (45% vs 20%; P = 0.037) [39]. Similar results tions and does not interact with the cytochrome 450 enzyme
were recently published by Cai et al. [40], who found that system. The RE-LY trial, which included 18,113 patients,
only metabolic syndrome and obesity, but not diabetes, were showed that dabigatran administered at a dose of 150 mg,
independent predictors of late AF recurrence (> 3 months compared with warfarin, was associated with lower rates
after ablation). of stroke and systemic embolism but similar rates of major
A recent meta-analysis has shown that valvular AF haemorrhage in patients with AF and at least one additional
and left atrial enlargement (diameter > 50 mm) were the risk factor, with a similar risk of major bleeding [44]. The
strongest predictors of AF recurrence [41]. Diabetes, how- same study demonstrated that diabetes was associated with
ever, was not one of the independent predictors. major bleeding risk in patients treated with dabigatran [45].

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
6 M. Tadic, C. Cuspidi

Among direct factor Xa inhibitors, apixaban seems to patients with risk factors for AF (hypertension, diabetes,
be the most promising and is approved for stroke preven- coronary artery disease and heart failure) revealed that the
tion. The ROCKET AF trial demonstrated that rivaroxaban use of ACE inhibitors or ARBs could not prevent recurrence
was non-inferior to warfarin for the prevention of stroke of AF after catheter ablation [58]. On the other hand, Fogari
or systemic embolism in patients with AF. There was et al. showed that valsartan was effective, even more than
no significant difference in the risk of major bleeding, atenolol, in preventing recurrent paroxysmal AF (20.3% vs
although intracranial and fatal bleedings were less fre- 34.1%) in 296 patients with hypertension and diabetes [59].
quent in the rivaroxaban group [46]. The same trial showed Some non-antiarrhythmic drugs directed against atrial
that diabetes was not an independent risk factor of bleed- structural and electrical remodelling, inflammation and
ing in patients with AF treated with rivaroxaban [47]. oxidative stress have been suggested as novel therapeutic
Furthermore, a recently published study that included approaches in the management of AF. This group of agents
27,467 patients treated with rivaroxaban demonstrated that includes statins, n-3 polyunsaturated fatty acids and several
diabetes was not more prevalent among patients who expe- antioxidant drugs such as vitamins C and E, N-acetylcysteine
rienced major bleeding [48]. The latest analyses of the and xanthine oxidase inhibitors [60—62]. However, we still
ARISTOTLE study found that diabetes was associated with need prospective studies to confirm the protective effect of
major bleeding in patients with AF receiving apixaban these agents on AF occurrence in patients with diabetes.
[49,50]. A recent meta-analysis that included all major trials
regarding new anticoagulant drugs showed that these drugs
were more effective and safer than vitamin K antagonists Conclusions
in reducing stroke and embolic events or major bleeding in
diabetic patients with AF [51]. However, special attention AF is commonly present in patients with diabetes, and these
should be directed to reversion of atrial remodelling in dia- two conditions will be seen together more frequently in the
betic patients with AF. Insulin resistance is one of the most future because the prevalence of both is increasing. Thus, it
important targets in the management of these patients. Thi- is very important to establish the most effective treatment
azolidinediones belongs to a class of insulin-sensitive agents in this subpopulation of patients with AF. New prospective
that ameliorate insulin resistance in patients with diabetes. studies with large numbers of patients with diabetes and
Chao et al. researched the possible relationship between AF are needed to investigate the mechanisms of this rela-
thiazolidinedione treatment and development of new-onset tionship and all possible therapeutic approaches in order to
AF in 12,605 patients with diabetes [52]. During a follow- determine the best possible individual management of both
up of 5 years, thiazolidinediones reduced the risk of AF conditions.
occurrence by 31% after adjustment for age, underlying dis-
eases and baseline medications. Gu et al. demonstrated that
pioglitazone improved the preservation of sinus rhythm and Disclosure of interest
reduced the re-ablation rate in patients with diabetes who
underwent catheter ablation due to AF [53]. A meta-analysis The authors declare that they have no conflicts of interest
of 19 randomized clinical studies showed that pioglitazone concerning this article.
was associated with a significantly lower risk of death,
myocardial infarction and stroke among patients with dia- Funding: None.
betes [54].
Probucol, a lipid-lowering drug with a potent antioxi-
dant effect, could positively influence atrial remodelling in References
AF [55]. Specifically, this agent may reduce the production
of oxygen free radicals, increase the expression of antioxi- [1] International Diabetes Federation. Key findings; 2014 [Avail-
dant enzymes, such as glutathione peroxidase, prevent lipid able from: http://www.idf.org/diabetesatlas/update-2014 (2
peroxidation, and inhibit the activity of nicotinamide ade- February 2014)].
nine dinucleotide phosphate-oxidase. Gong et al. reported [2] Martin-Timon I, Sevillano-Collantes C, Segura-Galindo A, Del
Canizo-Gomez FJ. Type 2 diabetes and cardiovascular disease:
that probucol reduced C-reactive protein concentration and
have all risk factors the same strength? World J Diabetes
attenuated atrial oxidative stress, significantly decreased 2014;5:444—70.
atrial heterogeneous sympathetic hyper-innervation, and [3] Benjamin EJ, Levy D, Vaziri SM, et al. Independent risk fac-
radically reduced the inducibility and duration of AF [56]. tors for atrial fibrillation in a population-based cohort. The
Current data show that angiotensin-converting enzyme Framingham Heart Study. JAMA 1994;271:840—4.
(ACE) inhibitors and angiotensin II receptor blockers (ARBs) [4] Camm AJ, Kirchhof P, Lip GY, et al. Guidelines for the manage-
could prevent the development of AF. This could be due ment of atrial fibrillation: the Task Force for the management
to the up-regulation of the RAAS that is seen in diabetes, of atrial fibrillation of the European Society of Cardiology
and which represents a key mechanism in the pathogene- (ESC). Europace 2010;12:1360—420.
sis of AF. A meta-analysis of 26 randomized clinical trials [5] Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006
guidelines for the management of patients with atrial fibril-
estimating the effect of ACE inhibitors or ARBs on the pre-
lation: full text: a report of the American College of
vention of AF showed that both drugs had preventive effects Cardiology/American Heart Association Task Force on practice
on AF (odds ratio 0.65, 95% CI 0.55—0.76) and these effects guidelines and the European Society of Cardiology Committee
were similar in these two classes of drugs [57]. However, for Practice Guidelines (Writing Committee to Revise the 2001
ACE inhibitors and ARBs showed higher benefit effects on guidelines for the management of patients with atrial fibril-
recurrent AF than on new-onset AF. A subgroup analysis of lation) developed in collaboration with the European Heart

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
Diabetes and atrial fibrillation 7

Rhythm Association and the Heart Rhythm Society. Europace control study in 75-year old subjects. Cardiovasc Diabetol 2008;
2006;8:651—745. 7:28.
[6] Movahed MR, Hashemzadeh M, Jamal MM. Diabetes mellitus [26] Frost L, Hune LJ, Vestergaard P. Overweight and obesity as risk
is a strong, independent risk for atrial fibrillation and flut- factors for atrial fibrillation or flutter: the Danish Diet, Cancer,
ter in addition to other cardiovascular disease. Int J Cardiol and Health Study. Am J Med 2005;118:489—95.
2005;105:315—8. [27] Nichols GA, Reinier K, Chugh SS. Independent contribution of
[7] Sun Y, Hu D. The link between diabetes and atrial fibrillation: diabetes to increased prevalence and incidence of atrial fibril-
cause or correlation? J Cardiovasc Dis Res 2010;1:10—1. lation. Diabetes Care 2009;32:1851—6.
[8] Tayebjee MH, Lim HS, MacFadyen RJ, Lip GY. Matrix [28] Schoen T, Pradhan AD, Albert CM, Conen D. Type 2 diabetes
metalloproteinase-9 and tissue inhibitor of metalloproteinase- mellitus and risk of incident atrial fibrillation in women. J Am
1 and -2 in type 2 diabetes: effect of 1 year’s cardiovascular Coll Cardiol 2012;60:1421—8.
risk reduction therapy. Diabetes Care 2004;27:2049—51. [29] Ostgren CJ, Merlo J, Rastam L, Lindblad U. Atrial fibrillation
[9] Lim HS, MacFadyen RJ, Lip GY. Diabetes mellitus, the renin- and its association with type 2 diabetes and hypertension in a
angiotensin-aldosterone system, and the heart. Arch Intern Swedish community. Diabetes Obes Metab 2004;6:367—74.
Med 2004;164:1737—48. [30] Aksnes TA, Schmieder RE, Kjeldsen SE, et al. Impact of new-
[10] Zhang Q, Liu T, Ng CY, Li G. Diabetes mellitus and atrial onset diabetes mellitus on development of atrial fibrillation
remodeling: mechanisms and potential upstream therapies. and heart failure in high-risk hypertension (from the VALUE
Cardiovasc Ther 2014;32:233—41. Trial). Am J Cardiol 2008;101:634—8.
[11] Kato T, Yamashita T, Sekiguchi A, et al. What are arrhyth- [31] Huxley RR, Lopez FL, Folsom AR, et al. Absolute and
mogenic substrates in diabetic rat atria? J Cardiovasc attributable risks of atrial fibrillation in relation to optimal and
Electrophysiol 2006;17:890—4. borderline risk factors: the Atherosclerosis Risk in Communities
[12] Watanabe M, Yokoshiki H, Mitsuyama H, et al. Conduction and (ARIC) study. Circulation 2011;123:1501—8.
refractory disorders in the diabetic atrium. Am J Physiol Heart [32] Thacker EL, McKnight B, Psaty BM, et al. Association of body
Circ Physiol 2012;303:H86—95. mass index, diabetes, hypertension, and blood pressure lev-
[13] Chao TF, Suenari K, Chang SL, et al. Atrial substrate properties els with risk of permanent atrial fibrillation. J Gen Intern Med
and outcome of catheter ablation in patients with paroxysmal 2013;28:247—53.
atrial fibrillation associated with diabetes mellitus or impaired [33] Kerr CR, Humphries KH, Talajic M, et al. Progression to chronic
fasting glucose. Am J Cardiol 2010;106:1615—20. atrial fibrillation after the initial diagnosis of paroxysmal atrial
[14] Tadic M, Ilic S, Cuspidi C, et al. Left and right atrial phasic fibrillation: results from the Canadian Registry of Atrial Fibril-
function and deformation in untreated patients with predia- lation. Am Heart J 2005;149:489—96.
betes and type 2 diabetes mellitus. Int J Cardiovasc Imaging [34] Tsang TS, Barnes ME, Miyasaka Y, et al. Obesity as a risk factor
2015;31:65—76. for the progression of paroxysmal to permanent atrial fibril-
[15] Kadappu KK, Boyd A, Eshoo S, et al. Changes in left atrial vol- lation: a longitudinal cohort study of 21 years. Eur Heart J
ume in diabetes mellitus: more than diastolic dysfunction? Eur 2008;29:2227—33.
Heart J Cardiovasc Imaging 2012;13:1016—23. [35] Latini R, Staszewsky L, Sun JL, et al. Incidence of atrial
[16] Mondillo S, Cameli M, Caputo ML, et al. Early detection of left fibrillation in a population with impaired glucose tolerance:
atrial strain abnormalities by speckle-tracking in hypertensive the contribution of glucose metabolism and other risk fac-
and diabetic patients with normal left atrial size. J Am Soc tors. A post-hoc analysis of the Nateglinide and Valsartan in
Echocardiogr 2011;24:898—908. Impaired Glucose Tolerance Outcomes Research trial. Am Heart
[17] Ayhan S, Ozturk S, Alcelik A, et al. Atrial conduction time and J 2013;166:935e1—40e1.
atrial mechanical function in patients with impaired fasting [36] Fontes JD, Lyass A, Massaro JM, et al. Insulin resistance and
glucose. J Interv Cardiovasc Electrophysiol 2012;35:247—52 atrial fibrillation (from the Framingham Heart Study). Am J
[discussion 52]. Cardiol 2012;109:87—90.
[18] Kamenov ZA, Traykov LD. Diabetic autonomic neuropathy. Adv [37] Huxley RR, Filion KB, Konety S, Alonso A. Meta-analysis of
Exp Med Biol 2012;771:176—93. cohort and case-control studies of type 2 diabetes mellitus and
[19] Dimitropoulos G, Tahrani AA, Stevens MJ. Cardiac autonomic risk of atrial fibrillation. Am J Cardiol 2011;108:56—62.
neuropathy in patients with diabetes mellitus. World J Diabetes [38] Forleo GB, Mantica M, De Luca L, et al. Catheter ablation of
2014;5:17—39. atrial fibrillation in patients with diabetes mellitus type 2:
[20] Oberhauser V, Schwertfeger E, Rutz T, Beyersdorf F, Rump results from a randomized study comparing pulmonary vein
LC. Acetylcholine release in human heart atrium: influence isolation versus antiarrhythmic drug therapy. J Cardiovasc Elec-
of muscarinic autoreceptors, diabetes, and age. Circulation trophysiol 2009;20:22—8.
2001;103:1638—43. [39] Chang SL, Tuan TC, Tai CT, et al. Comparison of out-
[21] Otake H, Suzuki H, Honda T, Maruyama Y. Influences of auto- come in catheter ablation of atrial fibrillation in patients
nomic nervous system on atrial arrhythmogenic substrates and with versus without the metabolic syndrome. Am J Cardiol
the incidence of atrial fibrillation in diabetic heart. Int Heart 2009;103:67—72.
J 2009;50:627—41. [40] Cai L, Yin Y, Ling Z, et al. Predictors of late recurrence
[22] Krahn AD, Manfreda J, Tate RB, Mathewson FA, Cuddy TE. of atrial fibrillation after catheter ablation. Int J Cardiol
The natural history of atrial fibrillation: incidence, risk fac- 2013;164:82—7.
tors, and prognosis in the Manitoba Follow-Up Study. Am J Med [41] D’Ascenzo F, Corleto A, Biondi-Zoccai G, et al. Which are
1995;98:476—84. the most reliable predictors of recurrence of atrial fibrilla-
[23] Dublin S, Glazer NL, Smith NL, et al. Diabetes mellitus, tion after transcatheter ablation? A meta-analysis. Int J Cardiol
glycemic control, and risk of atrial fibrillation. J Gen Intern 2013;167:1984—9.
Med 2010;25:853—8. [42] Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of
[24] Ruigomez A, Johansson S, Wallander MA, Rodriguez LA. Inci- the ESC Guidelines for the management of atrial fibrillation:
dence of chronic atrial fibrillation in general practice and its an update of the 2010 ESC Guidelines for the management
treatment pattern. J Clin Epidemiol 2002;55:358—63. of atrial fibrillation. Developed with the special contribu-
[25] Johansen OE, Brustad E, Enger S, Tveit A. Prevalence of tion of the European Heart Rhythm Association. Eur Heart J
abnormal glucose metabolism in atrial fibrillation: a case 2012;33:2719—47.

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009
+Model
ACVD-806; No. of Pages 8 ARTICLE IN PRESS
8 M. Tadic, C. Cuspidi

[43] Hart RG, Pearce LA, Rothbart RM, et al. Stroke with intermit- [52] Chao TF, Leu HB, Huang CC, et al. Thiazolidinediones
tent atrial fibrillation: incidence and predictors during aspirin can prevent new onset atrial fibrillation in patients with
therapy. Stroke Prevention in Atrial Fibrillation Investigators. non-insulin dependent diabetes. Int J Cardiol 2012;156:
J Am Coll Cardiol 2000;35:183—7. 199—202.
[44] Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran ver- [53] Gu J, Liu X, Wang X, et al. Beneficial effect of pioglitazone
sus warfarin in patients with atrial fibrillation. N Engl J Med on the outcome of catheter ablation in patients with paroxys-
2009;361:1139—51. mal atrial fibrillation and type 2 diabetes mellitus. Europace
[45] Reilly PA, Lehr T, Haertter S, et al. The effect of dabiga- 2011;13:1256—61.
tran plasma concentrations and patient characteristics on the [54] Lincoff AM, Wolski K, Nicholls SJ, Nissen SE. Pioglitazone and
frequency of ischemic stroke and major bleeding in atrial risk of cardiovascular events in patients with type 2 dia-
fibrillation patients: the RE-LY Trial (Randomized Evaluation betes mellitus: a meta-analysis of randomized trials. JAMA
of Long-Term Anticoagulation Therapy). J Am Coll Cardiol 2007;298:1180—8.
2014;63:321—8. [55] Liu T, Li G. Probucol and succinobucol in atrial fibrillation: pros
[46] Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban ver- and cons. Int J Cardiol 2010;144:295—6.
sus warfarin in nonvalvular atrial fibrillation. N Engl J Med [56] Gong YT, Li WM, Li Y, et al. Probucol attenuates atrial
2011;365:883—91. autonomic remodeling in a canine model of atrial fibrilla-
[47] Goodman SG, Wojdyla DM, Piccini JP, et al. Factors associ- tion produced by prolonged atrial pacing. Chin Med J (Engl)
ated with major bleeding events: insights from the ROCKET 2009;122:74—82.
AF trial (rivaroxaban once-daily oral direct factor Xa inhibi- [57] Zhang Y, Zhang P, Mu Y, et al. The role of renin-angiotensin
tion compared with vitamin K antagonism for prevention of system blockade therapy in the prevention of atrial fibrillation:
stroke and embolism trial in atrial fibrillation). J Am Coll Car- a meta-analysis of randomized controlled trials. Clin Pharmacol
diol 2014;63:891—900. Ther 2010;88:521—31.
[48] Tamayo S, Frank Peacock W, Patel M, et al. Characterizing [58] Tayebjee MH, Creta A, Moder S, et al. Impact of angiotensin-
major bleeding in patients with nonvalvular atrial fibrilla- converting enzyme-inhibitors and angiotensin receptor block-
tion: a pharmacovigilance study of 27,467 patients taking ers on long-term outcome of catheter ablation for atrial
rivaroxaban. Clin Cardiol 2015 [Epub ahead of print; doi: fibrillation. Europace 2010;12:1537—42.
10.1002/clc.22373]. [59] Fogari R, Zoppi A, Mugellini A, et al. Comparative evalua-
[49] Held C, Hylek EM, Alexander JH, et al. Clinical outcomes and tion of effect of valsartan/amlodipine and atenolol/amlodipine
management associated with major bleeding in patients with combinations on atrial fibrillation recurrence in hypertensive
atrial fibrillation treated with apixaban or warfarin: insights patients with type 2 diabetes mellitus. J Cardiovasc Pharmacol
from the ARISTOTLE trial. Eur Heart J 2014 [Epub ahead of 2008;51:217—22.
print; pii: ehu463]. [60] Liu T, Li L, Korantzopoulos P, Liu E, Li G. Statin use and
[50] Hylek EM, Held C, Alexander JH, et al. Major bleeding in development of atrial fibrillation: a systematic review and
patients with atrial fibrillation receiving apixaban or warfarin: meta-analysis of randomized clinical trials and observational
The ARISTOTLE Trial (apixaban for reduction in stroke and studies. Int J Cardiol 2008;126:160—70.
other thromboembolic events in atrial fibrillation): predic- [61] Liu T, Korantzopoulos P, Shehata M, et al. Prevention of atrial
tors, characteristics, and clinical outcomes. J Am Coll Cardiol fibrillation with omega-3 fatty acids: a meta-analysis of ran-
2014;63:2141—7. domised clinical trials. Heart 2011;97:1034—40.
[51] Lega JC, Bertoletti L, Gremillet C, et al. Consistency of safety [62] Liu T, Korantzopoulos P, Li G. Antioxidant therapies for
and efficacy of new oral anticoagulants across subgroups of the management of atrial fibrillation. Cardiovasc Diagn Ther
patients with atrial fibrillation. PLoS One 2014;9:e91398. 2012;2:298—307.

Please cite this article in press as: Tadic M, Cuspidi C. Type 2 diabetes mellitus and atrial fibrillation: From mechanisms
to clinical practice. Arch Cardiovasc Dis (2015), http://dx.doi.org/10.1016/j.acvd.2015.01.009