HLC 2386 1–8

Heart, Lung and Circulation (2017) xx, 1–8 REVIEW

1443-9506/04/$36.00
http://dx.doi.org/10.1016/j.hlc.2017.05.116

1
2 Atrial Fibrillation in Hypertrophic
3 Cardiomyopathy
4 Q1 Kaivan Vaidya a*, Christopher Semsarian a,b,c, Kim H. Chan a,b
5 Q2
Q3 a
Department of Cardiology, Royal Prince Alfred Hospital, Sydney, NSW, Australia
6 Q4 b
Sydney Medical School, University of Sydney, Sydney, NSW, Australia
7 c
Agnes Ginges Centre for Molecular Cardiology, Centenary Institute, Sydney, NSW, Australia
8
9

Q5 Hypertrophic cardiomyopathy (HCM) is an inherited cardiac disorder with a spectrum of clinical mani-
festations. Patients with HCM are predisposed to developing atrial fibrillation (AF) due primarily to
Q6 advanced diastolic dysfunction and left atrial (LA) dilatation and remodelling. Atrial fibrillation causes a
progressive symptomatic and functional decline, as well as increased thromboembolic risk and mortality,
particularly in the setting of rapid ventricular rates and left ventricular outflow tract (LVOT) obstruction.
The mainstay of management of AF in HCM is a combination of non-pharmacological lifestyle and risk
factor modification, long-term anticoagulation, and rhythm control with antiarrhythmic medications. There
is a growing body of evidence indicating that an early and aggressive rhythm control strategy may result in
more favourable outcomes.

10
11 HCM, with a focus on pathophysiology, predisposing fac- 32
12 Introduction tors and management. Q8 33
13 Q7 Hypertrophic cardiomyopathy (HCM) is clinically defined
14 as the presence of left ventricular (LV) hypertrophy, usually
15 asymmetric and involving the interventricular septum, in
Prevalence and Incidence 34

16 the absence of abnormal loading conditions. It is most often Hypertrophic cardiomyopathy is the most common inherited 35
17 caused by autosomal dominant mutations in one of several cardiac condition, with a population prevalence of up to 1 in 36
18 sarcomere genes which code for components of the contrac- 200 [3–6,10]. It is an important cause of heart failure, and is a 37
19 tile apparatus [1–7]. Hypertrophic cardiomyopathy is a het- common cause of unexplained sudden death in the young [11]. 38
20 erogeneous disease with a wide spectrum of clinical Atrial fibrillation is the most common sustained arrhythmia in 39
21 manifestations, including an asymptomatic state, symptom- both the general and HCM population, with rates of AF four- to 40
22 atic heart failure, arrhythmias, and sudden cardiac death. six-fold higher in patients with HCM than the similarly aged 41
23 Atrial fibrillation (AF) is the most common sustained general population [3]. Previous studies, including a meta- 42
24 arrhythmia in patients with HCM, and is known to be analysis of 7,381 patients, have estimated an annual incidence 43
25 associated with adverse clinical outcomes and a worse of 2–3%, and a lifetime prevalence of approximately 20–30% in 44
26 prognosis [2–6,8,9]. Although prevention of sudden death HCM patients, with rates as high as 40% in those older than 70 45
27 and management of LV outflow tract (LVOT) obstruction years [3–6,12]. In a retrospective study of 4,248 patients, all of 46
28 have traditionally been the key priorities of HCM manage- whom were in sinus rhythm at baseline, 740 (17.4%) developed 47
29 ment, AF is more common than both sudden death and AF within 10 years [2]. Atrial fibrillation tends to be paroxys- 48
30 medically refractory obstruction [3]. This review article mal in two-thirds of HCM patients and persistent/permanent 49
31 covers a range of issues concerning AF in patients with in the remaining one-third [13,14]. 50

*Corresponding author at: Department of Cardiology, Royal Prince Alfred Hospital, Missenden Road, Camperdown NSW 2050, Australia.,
Email: kaivan.vaidya@gmail.com
Crown Copyright © 2017 Published by Elsevier B.V. on behalf of Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac
Society of Australia and New Zealand (CSANZ). All rights reserved.

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

2 K. Vaidya et al.

Furthermore, patients with HCM have a higher prevalence 82

51 Pathophysiology of atrial fibrosis [6,17] and atrial myofibril disarray, which 83
52 The development of AF in HCM patients is likely a multi- can serve as an arrhythmogenic substrate for AF by impair- 84
53 factorial process, including genetic factors, structural abnor- ing conduction of sinus impulses and causing intra-atrial re- 85
54 malities, and electrophysiological abnormalities causing entry [6,8,9]. Finally, other proposed mechanisms for precip- 86
55 impaired atrial conduction. itating AF in HCM include abnormal calcium handling caus- 87
56 Mutations in sarcomeric proteins account for approxi- ing triggered activity from delayed after-depolarisations, 88
57 mately 60% of HCM cases, and of the genes known to be hypertrophy of the muscle sleeves responsible for conduct- 89
58 associated with HCM, the three most predominant are ing pulmonary vein triggers to the LA, and coronary micro- 90
59 b-myosin heavy chain (MYH7), cardiac troponin T, and vascular dysfunction which results in atrial ischaemia/ 91
60 myosin-binding protein C [7]. The mis-sense mutation infarction and provides a substrate for development of AF 92
61 Arg663His in the MYH7 gene has been associated with an [8,9,18]. 93
62 increased risk of AF (47% prevalence over a seven-year In patients with HCM, the presence of AF with a rapid 94
63 follow-up period) [15], and polymorphisms in the angioten- ventricular response reduces LV diastolic filling time and 95
64 sin receptor gene (AGTR1) have also been linked to the causes loss of organised atrial depolarisation and contraction 96
65 development of AF in patients with HCM [6]. during diastole. These factors, compounded by reduced LV 97
The diagnosis of HCM has been shown to precede the onset compliance in a hypertrophied LV, can result in reduction in 98
66 of AF in the majority of cases, suggesting that structural and cardiac output and subsequent clinical deterioration [4,6]. 99
67 electrophysiological abnormalities play the main role in path- Atrial fibrillation is, therefore, often poorly tolerated by 100
68 ogenesis [5,13] (Figure 1). In HCM, a hypertrophied LV with patients with HCM, with common symptoms being palpita- 101
69 reduced compliance results in impaired diastolic filling of tions, dyspnoea, and chest pain. Patients with concomitant 102
70 blood from the left atrium (LA). Therefore, an elevated LV LVOT obstruction are also predisposed to hypotension, pre- 103
71 end-diastolic pressure (LVEDP) due to diastolic dysfunction syncope, and syncope. In a study of 52 HCM patients [14], the 104
72 causes an increase in LA afterload, and the LA undergoes acute onset of AF caused worsening symptoms in 89% of 105
73 progressive dilatation causing a secondary left atrial myopa- patients, with 93% returning to their original symptom class 106
74 thy. This process of atrial stretch and remodelling is exacer- with reversion to sinus rhythm or ventricular rate control. 107
75 bated by the presence of a LVOT obstruction as well as
76 associated systolic anterior motion of the mitral valve causing
77 mitral regurgitation, both of which are common phenotypic Risk Factors for Development of 108
78 features of HCM [4,6]. Left atrial dilatation and remodelling
79 shortens the effective atrial refractory period, which, in turn,
AF in HCM 109

80 increases the dispersion of repolarisation, thus potentiating the The strongest independent predictors of AF in HCM patients 110
81 ability of ectopic triggers to maintain AF [9,16]. described in the literature are LA diameter and volume, age, 111

Figure 1 Pathophysiology of AF in HCM.

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

Atrial Fibrillation in Hypertrophic Cardiomyopathy 3

Q1 Table 1 Predictors of AF in patients with HCM.

Electrocardiography Strength of the association

Filtered P-wave duration [23] 140 ms OR 3.44 (p < 0.0001)

Maximum P-wave duration [19] >134.5 ms RR 9.9 (p = 0.001)
P-wave dispersion [19] >52.5 ms RR 24 (p = 0.001)
Imaging (2D echocardiography or CMR)
LA diameter [13] >45 mm OR 3.4 (p < 0.0001)
LA volume index [20] 34 mL/m2 Area under ROC curve = 0.836
LA dysfunction [21] 16% global LA fractional shortening OR 0.72 (p < 0.001)
Septal hypertrophy [22] >0.5 x [base-to-apex length of septum] mm OR 5.44 (p < 0.001)

Abbreviations: 2D two dimensional, CMR cardiac magnetic resonance, LA left atrial, OR odds ratio, RR relative risk, ROC receiver operating
characteristic. Note: CMR used for septal hypertrophy [22].

112 and New York Heart Association (NYHA) heart failure func- shorter mean mitral E-wave deceleration time (216 ms AF 152
113 tional class, with LA diameter and volume not surprisingly versus 231 ms non-AF). Patients with ejection fraction (EF) < 153
114 being the most strongly associated (Table 1). 50% were more likely to have AF (46%) versus those with 154
115 In a meta-analysis of 7,381 patients (in 33 studies), the normal EF (17%). From a symptom point of view, AF was 155
116 pooled overall LA diameter was 38.03 mm compared to associated with higher rates of NYHA Class III/IV function 156
117 45.37 mm in HCM patients with sinus rhythm compared (46% AF versus 39% non-AF), and patients with AF had 157
118 to AF [12]. One study reported a LA diameter of over significantly worse exercise tolerance on cardiopulmonary 158
119 42 mm to be predictive of AF with 96% sensitivity and testing. 159
120 81% specificity [19]. In a study of 480 HCM patients [13], a Various studies have identified other structural predictors 160
121 LA diameter greater than 45 mm represented the threshold based on cardiac imaging. In a study of 1,360 HCM patients 161
122 value associated with substantial risk of AF developing, and [22], a greater extent of septal hypertrophy (OR, 5.44) on 162
123 this has now been incorporated into screening guidelines echocardiography and/or cardiac magnetic resonance 163
124 (discussed in Detection section below) (Table 1). An increased (CMR) was a significant independent predictor of progres- 164
125 LA volume index (LAVI; LA volume divided by body surface sion to AF (Table 1). Patients with more septal hypertrophy 165
126 area) has also been shown to be significantly associated with also had a larger LA diameter and volume, and higher 166
127 developing paroxysmal AF in 141 patients, with a cutoff of markers of severe LV diastolic dysfunction (E/A and E/E’ 167
128 34 mL/m2[1_TD$IF] (Table 1) identifying patients at risk with a ratio). Finally, in a study of 87 HCM patients [16], 37 (43%) of 168
129 sensitivity and specificity of 80% and 73% respectively whom had AF, the mean extent of ventricular late gadolin- 169
130 [6,20]. Left atrial function can also serve as a reliable inde- ium enhancement (LGE) on CMR imaging was higher in 170
131 pendent predictor, with one study reporting a higher risk of those with AF (12.4%) compared with those without 171
132 AF (odds ratio [OR], 0.72) in HCM patients with LA dysfunc- (6.0%), with LA volume related to the extent of LGE. 172
133 tion (measured as 16% global LA fractional shortening) Electrocardiographic features indicative of abnormal atrial 173
134 [21]. Finally, age is a well-known predictor, and various depolarisation have also been assessed as predictive of AF in 174
135 studies have reported an age at diagnosis threshold ranging HCM patients (Table 1). A prospective study of 110 patients 175
136 from 40 years to >50 years as independently predictive of demonstrated that patients with signal-averaged P-wave 176
137 AF in HCM patients [6,13]. durations greater than 140 ms are prone to developing AF 177
138 In a recent retrospective study of 4,248 HCM patients [2], with 56% sensitivity and 83% specificity [23]. When com- 178
139 multivariate analysis demonstrated an association between bined with LA diameter >40 mm, the sensitivity increased to 179
140 new-onset AF and age (hazard ratio [HR], 1.23), LA diameter 93%. Another study reported that a maximum P-wave dura- 180
141 (HR 2.39), female sex (HR, 1.35), and NYHA class II (HR, tion of >134.5 ms separated HCM patients with and without 181
142 1.29) and III–IV (HR, 1.65) symptoms, but not LVOT gradient AF with a 92% sensitivity and 89% specificity [19]. P-wave 182
143 or maximal LV wall thickness. Furthermore, there was a dispersion (difference between maximum and minimum P- 183
144 linear relationship between LA diameter and AF risk above wave duration) of >52.5 ms separated patients with AF from 184
145 a diameter of 40 mm. In a large retrospective study of 3,673 controls with 96% sensitivity and 91% specificity. 185
146 HCM patients [5], 650 (18%) of whom had AF, the presence of Left ventricular outflow tract obstruction in HCM confers 186
147 AF was significantly associated with larger mean LAVI an adverse prognosis, but evidence varies regarding whether 187
148 (62 mL/m2 AF versus 45 mL/m2 non-AF), higher rates of it predisposes to the development of AF [6,22]. In a study of 188
149 moderate or severe mitral regurgitation (20% AF versus 15% 3,673 patients [5], AF was more common among patients 189
150 non-AF), and markers of LV diastolic dysfunction such a with non-obstructive HCM, with a median resting LVOT 190
151 higher mean medial E/e’ ratios (19 AF versus 17 non-AF) and gradient of 21 mmHg and 31 mmHg in the AF and non- 191

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

4 K. Vaidya et al.

192 AF groups, respectively. Furthermore, AF was less common their efficacy. However, the often episodic and paroxysmal 224
193 (16%) in patients with resting or labile LVOT obstruction nature of AF may require more prolonged monitoring either 225
194 versus those with no obstruction (22%). A possible explana- with event or implantable loop recorders to increase the 226
195 tion for these findings is that patients with non-obstructive detection rate, especially in patients with intermittent symp- 227
196 HCM had worse diastolic dysfunction, but echocardio- toms [28]. 228
197 graphic evaluation of this is difficult in the setting of AF.
198 In contrast, however, other studies have suggested that the
199 development of AF is correlated with a higher LVOT gradi- Clinical Outcomes 229
200 ent [6]. This discrepancy between studies may be due to the
The development of persistent or permanent AF in patients 230
201 dynamic nature of the LVOT gradient in HCM.
with HCM often heralds a relentless functional decline, and 231
202 Finally, HCM has been reported to be associated with a
is associated with high rates of symptomatic heart failure, 232
203 high prevalence of obstructive sleep apnoea (OSA) [24], and
systemic thromboembolic complications, and mortality 233
204 it has been well reported that OSA is associated with the
[1,4,5,12]. Unlike ventricular tachyarrhythmias, however, 234
205 presence of AF [25]. Furthermore, the adverse haemody-
AF is generally not associated with an increased risk of 235
206 namic changes in LV preload, afterload, and diastolic func-
sudden cardiac death. 236
207 tion in OSA all lead to increases in LVOT obstruction and
Atrial fibrillation is an independent predictor of overall 237
208 further symptomatic decline in patients with HCM. How-
mortality in HCM patients, conferring an up to four-fold 238
209 ever, it remains to be seen whether OSA treatment will
increase in the risk of death compared with sinus rhythm 239
210 reduce AF burden.
[5,13]. The two major contributors to this are death due to 240
heart failure and stroke. In a study of 480 patients with HCM 241
followed for a mean of 9.1 years, with 107 (22%) who devel- 242
211 Detection of AF in HCM oped AF, AF was associated with an increased HCM-related 243
212 Due to the high prevalence of AF in HCM, both American mortality (OR, 3.7) on multivariate analysis [13]. Patients 244
213 and European guidelines have recommended AF screening who developed AF at a younger age (age <50 years) carried 245
214 in clinical practice. In European guidelines [26], 48-hour an even worse prognosis. Patients with AF had a greater rate 246
215 ambulatory electrocardiogram monitoring is recommended (OR, 2.8) of progression to NYHA Class III–IV heart failure 247
216 every 6 to 12 months in HCM patients who are in sinus symptoms compared to non-AF patients, and during follow- 248
217 rhythm and have a LA anterior-posterior diameter up 84% of patients who developed AF experienced deterio- 249
218 45 mm (Class IIA recommendation) (Table 2). The Ameri- ration in functional status (especially those with LVOT 250
219 can guidelines are less stringent [27], suggesting that 24-hour obstruction). In another recent study of 4,248 HCM patients 251
220 ambulatory electrocardiogram monitoring should be consid- with a median follow-up of 5.4 years, 740 (17.4%) of whom 252
221 ered in adults with HCM to assess for asymptomatic AF developed AF, there was a higher proportion of patients with 253
222 (Class IIB recommendation). These recommendations are cardiovascular death (10.9% AF versus 4.9% non-AF; relative 254
223 based on expert opinion with no published data supporting risk [RR], 2.2) and non-cardiovascular death (5.9% AF versus 255

Table 2 Management of AF in HCM Summary Table.

Issue Recommendation [26,27] Comments

Screening for
asymptomatic AF
Patients who should
receive anticoagulation
Anticoagulation therapy
Preferred antiarrhythmic
medications
Role for catheter ablation
48-hour Holter monitoring every 6-12
months if LA diameter  45 mm (Class IIA).
All patients (Class I).
Warfarin to target INR 2.0-3.0 (Class I).
Amiodarone followed by disopyramide
(Class IIA).
Consider in patients with drug refractory
symptomatic AF, especially those without
severe LA enlargement and shorter AF
duration (Class IIA).
LA diameter is the only risk stratification parameter used
currently.
Use of the CHA2DS2-VASc (or other risk stratification)
tool is not validated in HCM patients.
An oral direct thrombin inhibitor or factor Xa inhibitor
can be used as second-line therapy.
In practice sotalol is often used first line, to avoid side
effects of amiodarone and disopyramide. Perhaps an early
ablation strategy may be useful in patients concerned
about long-term toxicities.
Success rate in HCM is lower than in the general
population, due to a less favourable underlying AF
substrate and advanced atrial myopathy.

Abbreviations: AF atrial fibrillation, LA left atrial.

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

Atrial Fibrillation in Hypertrophic Cardiomyopathy 5

256 3.2% non-AF; RR, 1.8) at 10 years in the AF group versus the inotropic effect can exacerbate the LVOT gradient [26,27,31]. A 308
257 non-AF group [2]. rate control strategy may also be required in patients who have 309
258 Several studies have also documented the increased risk of failed or not tolerated attempts towards rhythm control, and 310
259 systemic thromboembolic events in HCM patients with AF. patients who are unsuitable for an ablation procedure. When 311
260 In a meta-analysis of 7,381 patients (in 33 studies) [12], the pharmacological rate control is ineffective or not tolerated, 312
261 overall prevalence of thromboembolism in HCM patients AVN ablation with pacemaker insertion will provide definitive 313
262 with AF was 27.1%, and the incidence of thromboembolism control. 314
263 was 3.8% per year. In the study of 480 HCM patients [13], For the majority of HCM patients, AF results in a symp- 315
264 stroke risk was markedly increased in patients with AF (OR, tomatic decline necessitating a rhythm control strategy aim- 316
265 17.7), and ischaemic stroke was eight times more frequent ing for restoration and maintenance of sinus rhythm. Despite 317
266 (21% AF; 2.6% non-AF). Stroke risk in AF was unrelated to the lack of any randomised controlled trials evaluating anti- 318
267 the type of AF (paroxysmal versus persistent/permanent) or arrhythmic drug therapy or ablation procedures in the HCM 319
268 the number of paroxysms experienced. Finally, in a retro- population, guidelines favour antiarrhythmic medications 320
269 spective multi-centre study of 4,821 HCM patients [1], mul- initially to prevent long-term recurrence of AF [26,27,31]. 321
270 tivariate analysis revealed an association between Amiodarone (Class III antiarrhythmic) is the least proar-
271 thromboembolism and AF (HR 8.41), and there appeared rhythmic and most effective at preventing AF recurrence. 322
272 to be a linear relationship between thromboembolic risk However, a multitude of potential toxicities (including 323
273 and LA diameter up to 45–50 mm, at which point the risk interstitial pneumonitis, thyroid and liver dysfunction, photo- 324
274 rose exponentially with increasing diameter. sensitivity, ophthalmic complications, and peripheral neurop- 325
athies) limit its use. Another option is disopyramide (Class IA 326
antiarrhythmic), particularly in young patients, to avoid long- 327
275 Management term toxicities. As it can accelerate AVN conduction, disopyr- 328
amide must be initiated with a concomitant AVN blocking 329
276 Patients with HCM are frequently more symptomatic of AF, agent in the setting of AF, and close monitoring is required 330
277 and persistent AF may herald a relentless functional decline, during dose up-titration for clinically significant prolongation 331
278 especially in the presence of rapid ventricular rates and of the QT interval, and anticholinergic and proarrhythmic side 332
279 LVOT obstruction. This warrants a more aggressive effects. In addition, disopyramide should only be administered 333
280 approach of restoration and maintenance of sinus rhythm to patients with obstructive HCM, as its negative inotropic Q10 334
281 [3,26,27]. Furthermore, as patients with HCM and AF are at a effects in patients without LVOT obstruction can precipitate 335
282 higher risk of thromboembolic complications, the recom- heart failure [3,26,27,31]. Sotalol (Class III antiarrhythmic) is an 336
283 mended threshold for anticoagulation is much lower. The alternative to disopyramide, but is also proarrhythmic and 337
284 key aspects of AF management include aggressive lifestyle hence careful monitoring is required [31]. Although not as 338
285 and risk factor control, rate and/or rhythm control, and effective, sotalol is often used first line in practice due to 339
286 thromboembolic prophylaxis (Table 2). concerns regarding side effects of amiodarone or disopyra- 340
mide. Finally, flecainide (Class IC antiarrhythmic) is contra- 341
287 Non-pharmacological and [2_TD$IF]Cardiovascular indicated in patients with structural heart disease such as HCM 342
288 Risk Factor Management [32], and the clinical efficacy and safety of dronedarone in 343
289 Aggressive lifestyle and cardiovascular risk factor manage- patients with HCM remains to be established. 344
290 ment in treatment of AF is often overlooked, and plays a vital In a study of 52 HCM patients with 46 who developed AF 345
291 complementary role in reducing AF burden [29,30]. [14], sinus rhythm was restored in 29 (63%) with amiodarone, 346
292 Although studies have not specifically included patients with and amiodarone was associated with successful maintenance 347
293 HCM, we recommend that aggressive lifestyle, dietary, and (76% over 5.5 years) of sinus rhythm and fewer embolic 348
294 risk factor modification should be undertaken in all HCM events. In a small study of 30 HCM patients, sotalol improved 349
295 patients with AF. In addition, weight loss may also improve exercise tolerance and successfully suppressed supraventric- 350
296 sleep disordered breathing which may drive AF [25]. ular (86%) and ventricular arrhythmias (54%) [33]. Finally, a 351
retrospective study of 4,248 HCM patients demonstrated that 352
297 Pharmacological [3_TD$IF]Management beta blockers, calcium channel blockers, and disopyramide 353
298 To date, there are no randomised prospective trials of drug initially prevented AF recurrence for two to three years, but 354
299 therapy for AF in HCM patients, but the initial management of this protective effect diminished with time, likely due to 355
300 AF in HCM patients is similar to that of the general population disease progression [2]. Contrary to expectations, amiodar- 356
301 with some exceptions. In patients with severe symptoms or one therapy did not prevent AF recurrence in this study, 357
302 haemodynamic instability, urgent cardioversion is required. In possibly because a large proportion of patients taking amio- 358
303 asymptomatic or minimally symptomatic patients, initial ther- darone did not have a history of AF at first evaluation. 359
304 apy includes control of ventricular rate with atrioventricular
305 nodal (AVN) blockers such as beta blockers or non-dihydro- AF [4_TD$IF]Ablation 360
306 pyridine calcium channel blockers (verapamil or diltiazem). Several studies have evaluated the role of radiofrequency- 361
307 Q9 Digoxin is contraindicated in patients with HCM as its positive based catheter ablation (RFCA) in selected HCM patients 362

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

6 K. Vaidya et al.

363 with drug refractory symptomatic AF [3]. This therapy, patients without HCM after a single procedure (38.7% 376
364 incorporating a technique of pulmonary vein isolation HCM versus 49.8% controls; OR, 2.25) and after one or more 377
365 (PVI), aims to eradicate AF triggers and abnormal atrial procedures (51.8% HCM versus 71.2% controls; OR, 2.62) [9]. 378
366 substrate leading to maintenance of AF. In a study of 61 The median number of procedures was 1.4 in HCM patients 379
367 HCM patients undergoing catheter ablation, restoration of and 1.2 in controls, and antiarrhythmic drugs were more 380
368 sinus rhythm without AF recurrence was achieved in 67% of frequently needed (OR, 4.70) following ablation in patients 381
369 patients over 29 months follow-up [8]. Repeat procedures with HCM compared to controls to prevent AF recurrence. 382
370 were required in 52% and antiarrhythmic therapy was main- The lower success rates in the HCM population reflects the 383
371 tained in 54% of patients post-ablation. Left atrial size and AF challenging underlying AF substrate (particularly LA dilata- 384
372 duration were most predictive of AF recurrence. In a system- tion and remodelling) and advanced atrial myopathy favour- 385
373 atic review of 14 observational studies with a median follow- ing AF recurrence [8,9]. Figure 2 illustrates an 386
374 up of 1.8 years, all of which involved RFCA and PVI, freedom electroanatomic LA voltage map of a patient with persistent 387
375 from atrial tachyarrhythmias post-ablation was higher in AF and HCM, with more extensive posterior wall scarring 388

Figure 2 Electroanatomic voltage map of LA and apical 4-chamber echocardiography of patient with persistent AF and
HCM (A and B) show extensive area of low voltage along posterior LA wall (green/yellow colour) and moderate LA
dilatation. In comparison, images (C and D) of patient with persistent AF without HCM show relatively normal posterior LA
wall voltage (purple colour) despite a similar degree of LA enlargement.

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

Atrial Fibrillation in Hypertrophic Cardiomyopathy 7

389 compared to a patient with persistent AF without HCM. The regimens have not been evaluated in randomised trials, 445
390 risk of procedure-related adverse events was low [9], with six a Vitamin K antagonist such as warfarin is recommended 446
391 studies reporting no major complications and the rest report- as first-line, with a target International Normalised Ratio 447
392 ing low rates of thromboembolic complications and pulmo- (INR) of 2.0 to 3.0. When warfarin is poorly tolerated, 448
393 nary vein stenosis. In a more recent study of 79 patients with difficult to maintain in the therapeutic range, or when 449
394 HCM who underwent catheter ablation [34], however, major regular INR monitoring is not possible, an oral direct 450
395 complications occurred in 6%. thrombin inhibitor (dabigatran) or factor Xa inhibitor 451
396 In the systematic review [9], patients with HCM under- (rivaroxaban/apixaban) is recommended. Several studies 452
397 went ablation relatively late in the course of their disease have reported a lower incidence of stroke in HCM patients 453
398 (median 5.9 years after diagnosis of atrial tachyarrhythmia), with AF treated with warfarin compared to antiplatelet 454
399 and therefore the greater degree of remodelling predisposed agents or no treatment [1,3,12,13], with a study of 4,821 455
400 to procedural failure. Despite this, the success rate of 51.8% patients reporting a 54.8% risk reduction in thromboem- 456
401 after a median of 1.4 procedures suggests that treatment bolic events with warfarin treatment compared to without 457
402 earlier in the natural history of the disease, with a more [1]. In patients not suitable for long-term anticoagulation 458
403 favourable substrate, may be quite effective. Furthermore, due to bleeding risk, LAA exclusion procedures can be 459
404 sensitivity analyses suggested that left atrial size was the considered [39], although this has not been specifically 460
405 major predictor of procedural success, and outcomes in evaluated in the HCM population. 461
406 HCM patients with paroxysmal AF and non-dilated atria Although the CHA2DS2-VASc score is commonly used for
407 (reflecting less advanced disease) were comparable to the stroke risk stratification in AF [31], it is not validated in the 462
408 non-HCM population. This may have implications for an HCM population. Furthermore, in the aforementioned 463
409 early and aggressive rhythm control strategy using catheter study [1], it did not correlate with thromboembolic end- 464
410 ablation in HCM patients with drug refractory AF, in partic- points. Therefore, current guidelines advise that all HCM 465
411 ular, younger patients with a shorter duration of AF and less patients with even a single brief episode of AF should be 466
412 dilated atria, and those hoping to avoid long-term toxicities treated with long-term anticoagulation, given the high risk 467
413 of medications such as amiodarone. This approach is sup- of stroke and AF recurrence, even if sinus rhythm is restored 468
414 ported by recent studies showing that a longer diagnosis-to- [26]. 469
415 ablation time for persistent AF is associated with markers of
416 adverse LA structural remodelling and worse outcomes [35],
417 and that AF ablation results in significant reverse remodel- Conclusion 470
418 ling [36].
Atrial fibrillation is the most common sustained arrhythmia 471
419 Further studies are also required to evaluate whether
in patients with HCM, and originates due to a multitude of 472
420 radiofrequency energy is the optimal energy source (versus
mechanical and electrical atrial adverse remodelling pro- 473
421 other sources such as cryo-energy) and whether extensive
cesses, in particular progressive LA dilatation. Atrial fibril- 474
422 ablation beyond PVI (such as linear ablation lines, posterior
lation is associated with significant symptomatic and 475
423 wall isolation, and targeting complex fractionated electro-
functional decline, as well as increased thromboembolic risk 476
424 grams, rotors, ganglionic plexi, and non-PV triggers) is more
and mortality. An early and aggressive rhythm control 477
425 effective. However, a recent randomised trial of 589 patients
strategy in conjunction with long-term anticoagulation is 478
426 with persistent AF showed no difference in freedom from AF
recommended in patients with HCM and AF. These key 479
427 between PVI alone versus PVI plus linear or complex frac-
points are presented in Box 1. Finally, there remain many 480
428 tionated electrogram ablation, although other ablation strat-
unanswered questions, and further prospective studies spe- 481
429 egies such as targeting rotors, ganglionic plexi and posterior
cifically addressing the HCM population are required to 482
430 wall isolation were not assessed [37].
elucidate the best strategies for early detection and treat- 483
431 There is some limited data suggesting surgical ablation of
ment of AF. 484
432 AF in HCM is also effective and safe [3,38]. Patients with
433 obstructive HCM who are undergoing a surgical myectomy
434 can opt for an adjunctive surgical MAZE procedure for AF
435 ablation. In a study of 68 HCM patients with refractory AF Box 1
436 who underwent surgery, 51% had no documented recur-
Key points
437 rence after a single procedure at a mean follow-up of 35
438 months, but major complications occurred in 18% [34]. Exclu-
439 sion of the left atrial appendage (LAA) should also be con-  AF is often poorly tolerated in patients with HCM
440 sidered at time of surgery.  Anticoagulation is recommended in all patients
with HCM and AF regardless of CHA2DS2-VASc
441 Thromboembolic [5_TD$IF]6Prophylaxis score
442 Guidelines unanimously recommend long-term therapeutic  An early and aggressive rhythm control strategy
443 anticoagulation for patients with HCM and AF to prevent should be considered in patients with HCM and AF
444 systemic thromboembolism [26,27]. Although specific

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116
 HLC 2386 1–8

8 K. Vaidya et al.

[20] Tani T, Tanabe K, Ono M, Yamaguchi K, Okada M, Sumida T, et al. Left 550
485 Disclosures atrial volume and the risk of paroxysmal atrial fibrillation in patients 551
with hypertrophic cardiomyopathy. J Am Soc Echocardiogr 2004;17 552
486 There are no disclosures. (6):644–8. 553
[21] Losi MA, Betocchi S, Aversa M, Lombardi R, Miranda M, D’Alessandro 554
G, et al. Determinants of atrial fibrillation development in patients with 555
hypertrophic cardiomyopathy. Am J Cardiol 2004;94(7):895–900. 556
487 References [22] Park KM, Im SI, Kim EK, Lee SC, Park SJ, Kim JS, et al. Atrial Fibrillation 557
in Hypertrophic Cardiomyopathy: Is the Extent of Septal Hypertrophy 558
488 [1] Guttmann OP, Pavlou M, O’Mahony C, Monserrat L, Anastasakis A, Important? PLoS One 2016;11(6):e0156410. 559
489 Rapezzi C, et al. Prediction of thrombo-embolic risk in patients with [23] Cecchi F, Montereggi A, Olivotto I, Marconi P, Dolara A, Maron BJ. Risk 560
490 hypertrophic cardiomyopathy (HCM Risk-CVA). Eur J Heart Fail 2015;17 for atrial fibrillation in patients with hypertrophic cardiomyopathy 561
491 (8):837–45. assessed by signal averaged P wave duration. Heart 1997;8(1):44–9. 562
492 [2] Guttmann OP, Pavlou M, O’Mahony C, Monserrat L, Anastasakis A, [24] Eleid MF, Konecny T, Orban M, Sengupta PP, Somers VK, Parish JM, 563
493 Rapezzi C, et al. Predictors of atrial fibrillation in hypertrophic cardio- et al. High prevalence of abnormal nocturnal oximetry in patients with 564
494 myopathy. Heart 2016;103(9):672–8. hypertrophic cardiomyopathy. J Am Coll Cardiol 2009;54(19):1805–9. 565
495 [3] MacIntyre C, Lakdawala NK. Management of Atrial Fibrillation in [25] Chan KH, Wilcox I. Obstructive sleep apnea: novel trigger and potential 566
496 Hypertrophic Cardiomyopathy. Circulation 2016;133(19):1901–5. therapeutic target for cardiac arrhythmias. Expert Rev Cardiovasc Ther 567
497 [4] Nair AG, Fischer AG. Atrial fibrillation in hypertrophic cardiomyopathy: 2010;8(7):981–94. 568
498 mechanisms, embolic risk and prognosis. Anadolu Kardiyol Derg 2006;6 [26] Elliott PM, Anastasakis A, Borger MA, Borggrefe M, Cecchi F, Charron P, 569
499 (Suppl 2):40–3. et al. 2014 ESC Guidelines on diagnosis and management of hypertrophic 570
500 [5] Siontis KC, Geske JB, Ong K, Nishimura RA, Ommen SR, Gersh BJ. Atrial cardiomyopathy: the Task Force for the Diagnosis and Management of 571
501 fibrillation in hypertrophic cardiomyopathy: prevalence, clinical corre- Hypertrophic Cardiomyopathy of the European Society of Cardiology 572
502 lations, and mortality in a large high-risk population. J Am Heart Assoc (ESC). Eur Heart J 2014;35(39):2733–79. 573
503 2014;3(3):e001002. [27] Gersh BJ, Maron BJ, Bonow RO, Dearani JA, Fifer MA, Link MS, et al. 2011 574
504 [6] Tuluce K, Tuluce SY. Predictors of Atrial Fibrillation Risk in Hypertro- ACCF/AHA guideline for the diagnosis and treatment of hypertrophic 575
505 phic Cardiomyopathy. J Atr Fibrillation 2015;7(5):1200. cardiomyopathy: a report of the American College of Cardiology Foun- 576
506 [7] Maron BJ, Maron MS, Semsarian C. Genetics of hypertrophic cardiomy- dation/American Heart Association Task Force on Practice Guidelines. J 577
507 opathy after 20 years: clinical perspectives. J Am Coll Cardiol 2012;60 Thorac Cardiovasc Surg 2011;142(6):e153–203. 578
508 (8):705–15. [28] Seet RC, Friedman PA, Rabinstein AA. Prolonged rhythm monitoring for 579
509 [8] Di Donna P, Olivotto I, Delcre SD, Caponi D, Scaglione M, Nault I, et al. the detection of occult paroxysmal atrial fibrillation in ischemic stroke of 580
510 Efficacy of catheter ablation for atrial fibrillation in hypertrophic cardio- unknown cause. Circulation 2011;124(4):477–86. 581
511 myopathy: impact of age, atrial remodelling, and disease progression. [29] Abed HS, Wittert GA, Leong DP, Shirazi MG, Bahrami B, Middeldorp 582
512 Europace 2010;12(3):347–55. ME, et al. Effect of weight reduction and cardiometabolic risk factor 583
513 [9] Providencia R, Elliott P, Patel K, McCready J, Babu G, Srinivasan N, management on symptom burden and severity in patients with atrial 584
514 et al. Catheter ablation for atrial fibrillation in hypertrophic cardiomy- fibrillation: a randomized clinical trial. JAMA 2013;310(19):2050–60. 585
515 opathy: a systematic review and meta-analysis. Heart 2016;102 [30] Pathak RK, Middeldorp ME, Meredith M, Mehta AB, Mahajan R, Wong 586
516 (19):1533–43. CX, et al. Long-Term Effect of Goal-Directed Weight Management in an 587
517 [10] Semsarian C, Ingles J, Maron MS, Maron BJ. New perspectives on the Atrial Fibrillation Cohort: A Long-Term Follow-Up Study (LEGACY). J 588
518 prevalence of hypertrophic cardiomyopathy. J Am Coll Cardiol 2015;65 Am Coll Cardiol 2015;65(20):2159–69. 589
519 (12):1249–54. [31] Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, et al. 590
520 [11] Bagnall RD, Weintraub RG, Ingles J, Duflou J, Yeates L, Lam L, et al. A 2016 ESC Guidelines for the management of atrial fibrillation developed 591
521 Prospective Study of Sudden Cardiac Death among Children and Young in collaboration with EACTS. Eur Heart J 2016;37(38):2893–962. 592
522 Adults. N Engl J Med 2016;374(25):2441–52. [32] Aliot E, Capucci A, Crijns HJ, Goette A, Tamargo J. Twenty-five years in 593
523 [12] Guttmann OP, Rahman MS, O’Mahony C, Anastasakis A, Elliott PM. the making: flecainide is safe and effective for the management of atrial 594
524 Atrial fibrillation and thromboembolism in patients with hypertrophic fibrillation. Europace 2011;13(2):161–73. 595
525 cardiomyopathy: systematic review. Heart 2014;100(6):465–72. [33] Tendera M, Wycisk A, Schneeweiss A, Polonski L, Wodniecki J. Effect of 596
526 [13] Olivotto I, Cecchi F, Casey SA, Dolara A, Traverse JH, Maron BJ. Impact sotalol on arrhythmias and exercise tolerance in patients with hypertro- 597
527 of atrial fibrillation on the clinical course of hypertrophic cardiomyopa- phic cardiomyopathy. Cardiology 1993;82(5):335–42. 598
528 thy. Circulation 2001;104(21):2517–24. [34] Bassiouny M, Lindsay BD, Lever H, Saliba W, Klein A, Banna M, et al. 599
529 [14] Robinson K, Frenneaux MP, Stockins B, Karatasakis G, Poloniecki JD, Outcomes of nonpharmacologic treatment of atrial fibrillation in patients 600
530 McKenna WJ. Atrial fibrillation in hypertrophic cardiomyopathy: a lon- with hypertrophic cardiomyopathy. Heart Rhythm 2015;12(7):1438–47. 601
531 gitudinal study. J Am Coll Cardiol 1990;15(6):1279–85. [35] Hussein AA, Saliba WI, Barakat A, Bassiouny M, Chamsi-Pasha M, Al- 602
532 [15] Gruver EJ, Fatkin D, Dodds GA, Kisslo J, Maron BJ, Seidman JG, et al. Bawardy R, et al. Radiofrequency Ablation of Persistent Atrial Fibrilla- 603
533 Familial hypertrophic cardiomyopathy and atrial fibrillation caused by tion: Diagnosis-to-Ablation Time, Markers of Pathways of Atrial Remod- 604
534 Arg663His beta-cardiac myosin heavy chain mutation. Am J Cardiol eling, and Outcomes. Circ Arrhythm Electrophysiol 2016;9(1):e003669. 605
535 1999;83(12A):13H–8H. [36] Walters TE, Nisbet A, Morris GM, Tan G, Mearns M, Teo E, et al. 606
536 [16] Papavassiliu T, Germans T, Fluchter S, Doesch C, Suriyakamar A, Haghi Progression of atrial remodeling in patients with high-burden atrial 607
537 D, et al. CMR findings in patients with hypertrophic cardiomyopathy fibrillation: Implications for early ablative intervention. Heart Rhythm 608
538 and atrial fibrillation. J Cardiovasc Magn Reson 2009;11:34. 2016;13(2):331–9. 609
539 [17] Ohtani K, Yutani C, Nagata S, Koretsune Y, Hori M, Kamada T. High [37] Verma A, Jiang CY, Betts TR, Chen J, Deisenhofer I, Mantovan R, et al. 610
540 prevalence of atrial fibrosis in patients with dilated cardiomyopathy. J Approaches to catheter ablation for persistent atrial fibrillation. N Engl J 611
541 Am Coll Cardiol 1995;25(5):1162–9. Med 2015;372(19):1812–22. 612
542 [18] Alasady M, Shipp NJ, Brooks AG, Lim HS, Lau DH, Barlow D, et al. [38] Chen MS, McCarthy PM, Lever HM, Smedira NG, Lytle BL. Effectiveness 613
543 Myocardial infarction and atrial fibrillation: importance of atrial ische- of atrial fibrillation surgery in patients with hypertrophic cardiomyopa- 614
544 mia. Circ Arrhythm Electrophysiol 2013;6(4):738–45. thy. Am J Cardiol 2004;93(3):373–5. 615
545 [19] Ozdemir O, Soylu M, Demir AD, Topaloglu S, Alyan O, Turhan H, et al. [39] Holmes DR, Reddy VY, Turi ZG, Doshi SK, Sievert H, Buchbinder M, 616
546 P-wave durations as a predictor for atrial fibrillation development in et al. Percutaneous closure of the left atrial appendage versus warfarin 617
547 patients with hypertrophic cardiomyopathy. Int J Cardiol 2004;94(2– therapy for prevention of stroke in patients with atrial fibrillation: a 618
548 3):163–6. randomised non-inferiority trial. Lancet 2009;374(9689):534–42. 619
549

Please cite this article in press as: Vaidya K, et al. Atrial Fibrillation in Hypertrophic Cardiomyopathy. Heart, Lung and
Circulation (2017), http://dx.doi.org/10.1016/j.hlc.2017.05.116