Download 1 - Europace

CLINICAL RESEARCH
Europace (2012) 14, 380–387
doi:10.1093/europace/eur305
Electrophysiology and Ablation
Prediction of sinus node dysfunction
in patients with persistent atrial flutter using
the flutter cycle length
Akinori Sairaku *, Yukiko Nakano, Noboru Oda, Yuko Makita, Kenta Kajihara,
Takehito Tokuyama, Chikaaki Motoda, Mai Fujiwara, and Yasuki Kihara
Department of Cardiology, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 7348551, Japan
Received 16 June 2011; accepted after revision 22 August 2011; online publish-ahead-of-print 16 September 2011
Aims
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Ablation † Atrial flutter † Flutter cycle length † Permanent pacemaker † Sinus node dysfunction † Atrial
remodelling
Introduction
Atrial flutter (AFL) is the second most common atrial tachyarrhythmia after atrial fibrillation. Catheter ablation (CA) of typical
AFL has been well established and ensures a high success rate
and adequate safety.1 However, sinus node dysfunction (SND)
occasionally becomes obvious when long-term persistent AFL is
terminated by CA,2,3 and a permanent pacemaker implantation
(PMI) is sometimes necessary. Therefore, because the sinus node
function cannot be assessed during AFL, preoperative prediction
of underlying SND may be helpful for patients; however, there is
no available data on this issue.
The atrial flutter cycle length (FCL) is one of the few electrophysiological (EP) parameters regarding the atria obtained from
the surface electrocardiogram (ECG) during AFL. Based on the
well-recognized ‘circus movement theory,4 the FCL is supposed
to depend on the atrial conduction velocity as well as the length
of the circuit. Hence, given that slowing of the conduction velocity
is closely related to SND in a diseased atria and provides an
arrhythmogenic substrate causing atrial tachyarrhythmias including
AFL,5 – 8 it is inferred that the FCL may reflect the sinus node function. The aim of the present study was to test the hypothesis that
the sinus node function can be predicted by assessing the FCL in
patients with persistent AFL.
* Corresponding author. Tel: +81 82 257 5540; fax: +81 82 257 5169, Email: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2011. For permissions please email: [email protected].
Downloaded from by guest on October 24, 2016
Sinus node dysfunction (SND) occasionally coexists with atrial flutter (AFL). However, the identification of SND
during AFL is difficult. We investigated whether we could predict underlying SND in patients with persistent AFL
using the flutter cycle length (FCL).
.....................................................................................................................................................................................
Methods
We retrospectively studied 211 successfully ablated patients with persistent cavotricuspid isthmus (CTI)-dependent
AFL and measured the FCL before the ablation and corrected sinus node recovery time (CSNRT) after the ablation.
and results
Twenty-four patients (11%) required a permanent pacemaker implantation (PMI) for significant SND after AFL termination and had a longer FCL (295 + 37 vs. 236 + 34 ms; P , 0.0001) and greater CSNRT (1727 + 1014 vs.
603 + 733 ms; P , 0.0001) than those not requiring a PMI. A receiver-operating characteristic curve identified an
FCL of .273 ms as the optimal cut-off value for predicting SND requiring a PMI (area under the curve 0.91; sensitivity, 83% and specificity, 89%; P , 0.0001). Multiple linear and logistic regression analyses revealed that the left ventricular ejection fraction (LVEF) (b ¼ – 0.2; P ¼ 0.0016) and FCL (b ¼ 0.46; P , 0.0001) were independently
associated with the CSNRT, and that females [odds ratio (OR), 2.43; 95% confidence interval (CI), 1.32– 4.62;
P ¼ 0.0046], an LVEF , 50% (OR, 2.10; 95% CI, 1.20–3.87; P ¼ 0.012), and an FCL of .273 ms (OR, 5.34; 95%
CI, 3.08 –10.08; P , 0.0001) were independent predictors of SND requiring a PMI.
.....................................................................................................................................................................................
Conclusion
Although this study was based on a review of a database, the results suggest that assessing the FCL in patients with
persistent CTI-dependent AFL could be helpful in the risk stratification of underlying SND.
381
Prediction of SND in patients with persistent AFL using the FCL
Methods
Study population
Echocardiography
Prior to the EP study sessions, transthoracic echocardiography was
performed in all patients (iE 33, Philips Electronics Healthcare,
Bothell, WA, USA). The maximal left atrial diameter was measured
using B-mode ultrasonography. The left ventricular (LV) end-diastolic
volume (EDV) and end-systolic volume (ESV) were calculated from
the apical two- and four-chamber views applying a modified Simpson’s
method. Thereafter, the LV ejection fraction (LVEF) was calculated as
the LVEF ¼ (EDV – ESV)/EDV × 100.9
Electrophysiological study and catheter
ablation
All antiarrhythmic agents and digitalis were ceased ≥5 half-lives before
the EP procedure, and a signed consent form was obtained from each
patient. The EP study and CA were performed by a standard method,
as previously described.10 A Duo-decapolar catheter with 2 – 5 – 2 mm
interelectrode spacing (Duo-decapolar catheter; St. Jude Medical,
St. Paul, MN, USA) was positioned in the right atrium (RA), parallel
to the TA so that the distal pole was located in the medial region of
the CTI. A decapolar catheter with 4 mm interelectrode spacing
(St. Jude Medical) was inserted within the coronary sinus (CS), with
the proximal bipole located at its ostium. A pentapolar catheter with
4 mm interelectrode spacing (Medtronic, Minneapolis, MN, USA)
was advanced to the His bundle position. Then, a non-irrigated ablation catheter with a 4 mm tip (Navistar; Biosense Webster,
Diamond Bar, CA, USA) was placed close to the right ventricle (RV)
within the CTI. Bipolar intracardiac electrograms that were filtered
between 30 and 500 Hz were digitally recorded and stored simultaneously with a 12-lead surface ECG.
The FCL was preliminarily calculated on the surface ECG the day
before the ablation procedure by averaging 30 consecutive cycles in
lead V1. In the same fashion, the FCL was measured at the proximal
CS 10 min after the insertion of the sheaths without any sedation so
Pacemaker implantation
The patients received temporary pacing from a catheter within the CS
if they manifested any significant SND following AFL termination by
the CA. The indication for the temporary pacing included one or
more of the following: (i) sinus bradycardia with a heart rate of ,50
beats/min, (ii) sinus arrest, or (iii) repeated sinus pauses longer than
3 s. The patients who had bradycardia-related symptoms resulting
from the above EP abnormalities even 2 weeks after the CA had permanent dual-chamber pacemakers implanted.
Statistical analysis
Statistical analyses were performed using JMP software version 8.0
(SAS Institute, Tokyo, Japan). The clinical characteristics of the study
subjects are presented as percentages for categorical variables and
Downloaded from by guest on October 24, 2016
The study was approved by the institutional research board of the Hiroshima University Hospital and the hospital’s ethics committee. We
retrospectively reviewed the electrophysiology laboratory database
at Hiroshima University Hospital from 1999 to 2009 to identify the
patients who had undergone CA for cavotricuspid isthmus (CTI)dependent AFL. Eligible criteria included the following: (i) cavotricuspid
isthmus participation in the arrhythmic circuit confirmed by an EP
study, (ii) an atrial activation pattern during atrial tachycardia (AT)
showing a clockwise or counterclockwise rotation around the tricuspid
annulus (TA), (iii) termination of the AFL by a CTI linear ablation, and
(iv) atrial flutter persisting for more than 2 weeks. Typical AFL was
diagnosed when the surface ECG showed flutter waves that were predominantly negative in leads II, III, and aVF and positive in lead V1, with
a regular atrial rate.1 Patients with an uncharacteristic surface ECG of
typical AFL were also considered eligible for enrolment as well as
those with typical AFL if their atrial tachyarrhythmias were confirmed
to have arrhythmic circuits including the CTI and TA. However,
patients with atypical flutter waves were not included in our study if
their flutter wave amplitude was less than 0.1 mV in lead V1 of the
surface ECG. Also, patients were excluded if they had previously
undergone an AFL ablation or implantation of a pacemaker for SND
and if it was not possible to assess their sinus node function because
of failing to return to sinus rhythm within 20 min after the termination
of the AFL.
that the FCL to be calculated would not be affected by any possible
changes in the autonomic tone resulting from a venipuncture or sedative drugs. The ventricular cycle length during AFL was measured on
the surface ECG also in the same manner. Cavotricuspid isthmus
dependence was confirmed if concealed entrainment was identified
when pacing the CTI,11,12 and if the difference between the post-pacing
interval at the CTI and FCL was within 30 ms.11 A counterclockwise or
clockwise activation sequence around the TA during AFL was confirmed by sequential mapping in patients with typical flutter waves
and by both sequential mapping and electroanatomic mapping with a
CARTO system (Biosense Webster) in patients with atypical flutter
waves. Subsequently, a linear lesion was made by continuously applying
radiofrequency energy with a temperature target of 608C and power
limit of 50 W during a stepwise withdrawal of the ablation catheter
from the RV towards the inferior vena cava. Procedural success was
defined as the termination of AFL during the radiofrequency application and the creation of a bidirectional CTI block.
Twenty minutes after the successful AFL ablation, the baseline intracardiac conduction intervals including the sinus rhythm cycle length,
atrio-His (AH) and His-ventricle (HV) intervals were measured
during sinus rhythm. The conduction time along the TA was measured
from the distal bipole 1 – 2 to proximal bipole 19– 20 of a Duodecapolar catheter during pacing from bipole 1– 2. The conduction
time along the CS was measured from the proximal bipole 9 –10 to
the distal bipole 1 – 2 of the CS catheter during pacing from bipole
9 – 10. The conduction was measured at pacing cycle lengths of
500 ms after stable capture for at least 10 s, and the average conduction time per beat was used for the analyses. The sinus node recovery
time (SNRT), evaluated by 30 s burst pacing trains delivered every
50 ms from 600 to 300 ms, was determined as the longest time
from the stimulus artifact to the earliest atrial activity: a sinus node
impulse arriving in the atria confirmed by comparing the P-wave morphology of the surface ECG during sinus rhythm obtained previously.
The corrected SNRT (CSNRT) was determined by correcting for
the underlying sinus cycle length.13 The anterograde atrioventricular
(AV) nodal effective refractory period (AVNERP) was measured
using an eight-beat drive at a cycle length equal to the sinus cycle
length minus 100 ms, followed by a single premature atrial stimulus
introduced decrementally at 10 ms intervals. The AVNERP was
defined as the longest coupled premature atrial stimulus interval that
failed to propagate to the His bundle. Dual AV nodal physiology was
diagnosed if there was an AH jump, defined as an increment of
50 ms or more in the AH value for a decrement in the pacing cycle
length of 10 ms during the stimulation protocol. Multiple ATs were
diagnosed when more than one sustained AT besides a CTI-dependent
AFL was induced during atrial pacing.
382
A. Sairaku et al.
all of them. Of those 14 patients, 10 had a prior cardiac surgery
including 9 with valve replacements, 4 had non-ischaemic cardiomyopathy, and 2 had a clockwise activation sequence during
AFL. Overall, 32 (15%) patients were temporally paced for the
manifestation of SND following the termination of the AFL, and
24 (11%) had a permanent pacemaker implanted later for symptomatic SND; 7 patients for significant sinus bradycardia, 4 for sinus
arrest with brady escape beats and 13 for repetitive sinus pauses.
The patient baseline characteristics with or without SND requiring
a PMI are summarized in Table 1. Patients with SND requiring a
PMI were more likely to be women (42 vs. 15%; P ¼ 0.0014),
have prior cardiac surgery (33 vs. 12%; P ¼ 0.0061), structural
heart disease (54 vs. 29%; P ¼ 0.012), and a lower LVEF (54 +
11 vs. 61 + 12%; P ¼ 0.01). As for the medications, there was
no difference in the antiarrhythmic agents or digitalis prescribed,
and no patients were administered amiodarone.
Results
Electrophysiological parameters after
termination of the atrial flutter
Patient characteristics
Table 2 represents the EP parameters in the patients with or
without SND requiring a PMI. The patients with a PMI were
more likely to have atypical flutter waves (25 vs. 4%; P ¼
0.0001), longer FCLs measured from the proximal CS (295 + 37
vs. 236 + 34 ms; P , 0.0001), prolonged FCLs calculated preliminarily from the surface ECG on a different day (297 + 35 vs.
234 + 31 ms; P , 0.0001), longer sinus rhythm cycle lengths
(953 + 398 vs. 848 + 172 ms; P ¼ 0.049), longer conduction
times along the TA and CS (along TA; 116 + 28 vs. 85 + 20 ms,
P , 0.0001, along CS; 42 + 10 vs. 33 + 5 ms, P ¼ 0.0002),
A total of 228 patients were considered eligible for inclusion, but
the following 17 patients were excluded from our study; 5 patients
with a previous AFL ablation, 5 with a previous PMI for SND, 4 that
did not undergo EP testing for the absence of sinus recovery after
termination of the AFL, and 3 with atypical flutter waves with an
extremely low amplitude. Finally, 211 patients were enroled in
our study. The mean age of the enrolled patients was 66 + 29
years, 18% of which were women. Fourteen patients presented
with atypical flutter waves and CARTO mapping was applied in
Table 1 Clinical characteristics of the study patients
Variable
Pacemaker (2) (n 5 187)
Pacemaker (1) (n 5 24)
P value
...............................................................................................................................................................................
Age (years)
Female
66 + 11
28 (15%)
68 + 13
10 (42%)
0.37
0.0014
Duration of AFL (months)
2.6 + 4.8
2.1 + 3.1
0.59
23 (12%)
54 (29%)
8 (33%)
13 (54%)
Prior cardiac surgery
Structural heart disease
0.0061
0.012
Valvular heart disease
16 (9%)
6 (25%)
0.013
Ischaemic heart disease
Non-ischaemic cardiomyopathy
18 (10%)
16 (9%)
3 (13%)
4 (17%)
0.66
0.2
Hypertension
75 (40%)
12 (50%)
0.35
Diabetes mellitus
Antiarrhythmic drugs or digitalis
30 (16%)
96 (51%)
4 (17%)
13 (54%)
0.94
0.79
Class I antiarrhythmic drugs
65 (35%)
9 (38%)
0.79
Class IV antiarrhythmic drugs
b-blockers
29 (16%)
27 (14%)
2 (8%)
3 (13%)
0.35
0.79
Digitalis
2 (8%)
0.38
124 + 19
71 + 11
119 + 15
70 + 11
0.21
0.75
Left ventricular ejection fraction (%)
61 + 12
54 + 11
0.01
Left atrium diameter (mm)
39 + 7
41 + 7
0.19
Systolic blood pressure (mmHg)
Diastolic blood pressure (mmHg)
28 (15%)
Downloaded from by guest on October 24, 2016
means + SDs for continuous variables. The differences between the
patients with and without SND requiring a PMI were examined using
Pearson’s x2 tests for categorical variables or Student’s t-tests for continuous variables. Correlations between the clinical or EP parameters
were assessed by Pearson’s correlation test. Factors obtained before
termination of the AFL with a significant correlation to the CSNRT
were further analysed using a multiple linear regression analysis to
assess their independent relationship to the CSNRT. The ability of
the FCL and CSNRT to discriminate between patients with and
without SND requiring a PMI was evaluated by a receiver-operating
characteristic (ROC) curve analysis. The optimal cut-off values were
calculated by determining the FCL and CSNRT providing the greatest
sum of the sensitivity and specificity. Univariate and stepwise multivariate logistic regression analyses were performed using clinical and
EP parameters obtained before the termination of the AFL to determine the predictors of SND requiring a PMI. For all analyses, a
P-value of ,0.05 was considered statistically significant.
383
Prediction of SND in patients with persistent AFL using the FCL
Table 2 Pre- and post-ablation electrophysiological parameters
Variable
Pacemaker (– ) (n 5 187)
Pacemaker (1) (n 5 24)
P value
...............................................................................................................................................................................
Pre-ablation parameters
Atypical flutter waves
Flutter cycle length
8 (4%)
236 + 34
6 (25%)
295 + 37
0.0001
,0.0001
Ventricular cycle length during AFL
531 + 192
483 + 159
0.56
Post-ablation parameters
Sinus rhythm cycle length (ms)
848 + 172
953 + 398
0.049
85 + 20
116 + 28
,0.0001
33 + 5
1453 + 768
42 + 10
2700 + 985
0.0002
,0.0001
Corrected sinus node recovery time (ms)
603 + 733
1727 + 1014
,0.0001
AH interval (ms)
HV interval (ms)
104 + 47
50 + 11
173 + 196
56 + 15
0.0002
0.05
AV nodal effective refractory period (ms)
323 + 80
312 + 103
0.61
7 (29%)
0
0.21
0.22
Conduction time along the tricuspid annulus (ms)
Conduction time along the coronary sinus (ms)
Sinus node recovery time (ms)
Dual AV nodal pathways
Multiple atrial tachycardias
Time required to restore sinus rhythm after AFL termination (s)
29 (16%)
11 (6%)
24.5 + 125.3
290.5 + 386.0
,0.0001
AFL, atrial flutter; AV, atrioventricular; AH, atrio-His; HV, His-ventricle.
Downloaded from by guest on October 24, 2016
greater SNRT and CSNRT values (SNRT; 2700 + 985 vs. 1453 +
768 ms, P , 0.0001, CSNRT; 1727 + 1014 vs. 603 + 733 ms, P ,
0.0001), and required a longer time to return to sinus rhythm after
termination of the AFL (290.5 + 386.0 vs. 24.5 + 125.3 s; P ,
0.0001) than those without SND requiring a PMI (Table 2, Figure 1).
Correlations between the
electrophysiological parameters
A Pearson’s correlation coefficient test using the EP parameters
revealed that the FCL measured at the proximal CS was significantly correlated with the FCL calculated preliminarily on the
surface ECG (r ¼ 0.94; P , 0.0001), and the conduction
times along the TA (r ¼ 0.68; P , 0.0001) and CS (r ¼ 0.34;
P , 0.0001). The conduction times along the TA and CS in
turn were significantly correlated with the CSNRT (r ¼ 0.27;
P ¼ 0.0001, r ¼ 0.49; P , 0.0001, respectively). The CSNRT also
was significantly correlated to the time from the termination of
the AFL to the return to sinus rhythm (r ¼ 0.78; P , 0.0001).
Correlations between the pre-ablation
parameters and the corrected sinus node
recovery time
A Pearson’s correlation coefficient test using the parameters
obtained before the termination of the AFL showed that atypical
flutter waves (r ¼ 0.27; P ¼ 0.0001) and the FCL (r ¼ 0.5;
P , 0.0001) were significantly correlated with the CSNRT. A
multivariate linear regression analysis using the factors with a statistical significance including those parameters revealed that the
LVEF (b ¼ 20.2; P ¼ 0.0016) and FCL (b ¼ 0.46; P , 0.0001)
were independently associated with the CSNRT (Table 3).
Figure 1 Distribution of the flutter cycle lengths in patients
with or without sinus node dysfunction requiring a pacemaker
implantation.
Receiver-operating characteristic curve
analyses of the flutter cycle length to
predict the risk of sinus node dysfunction
requiring a pacemaker implantation
The ROC curve analyses showed that the FCL and CSNRT significantly discriminated between patients with and without SND
requiring a PMI with areas under the curve of 0.91 and 0.81
(both P , 0.0001), respectively (Figure 2). An FCL of .273 ms
and CSNRT of .1502 ms were identified as the optimal cut-off
values to predict SND requiring a PMI (sensitivity, 83 and 74%;
specificity, 89 and 95%; respectively).
384
A. Sairaku et al.
Table 3 Correlations between the corrected sinus
node recovery time and clinical or pre-ablation
electrophysiological parameters (n 5 211)
Variable
Univariate
Multivariate
r
b
........................ .......................
P value
P value
SND requiring a PMI. In the stepwise multivariate models, a
female gender [odds ratio (OR), 2.43; 95% confidence interval
(CI), 1.32–4.62; P ¼ 0.0046], LVEF ,50% (OR, 2.10; 95% CI,
1.20–3.87; P ¼ 0.012), and FCL of .273 ms (OR, 5.34; 95% CI,
3.08–10.08; P , 0.0001) were independent predictors of SND
requiring a PMI (Table 4).
................................................................................
Age
Female
20.0026
0.065
0.97
0.33
Duration of AFL
20.044
0.36
Prior cardiac surgery
Structural heart
disease
Diabetes mellitus
Antiarrhythmic agents
or digitalis
Systolic blood pressure
Diastolic blood
pressure
Left ventricular
ejection fraction
0.2
0.16
0.0044
0.0082
20.056
0.01
0.43
0.88
20.093
0.21
0.094
20.23
0.0014
0.028
0.71
Atypical flutter waves
Flutter cycle length
0.27
0.5
0.0001
,0.0001
0.55
0.19
Our major findings were as follows: (i) a longer FCL was significantly associated with a decreased sinus node function, and (ii) a
prolonged FCL was a strong predictor of SND requiring a PMI.
Association between atrial remodelling
and the flutter cycle length
20.2
0.033
0.46
0.0016
0.67
,0.0001
0.079
r, Pearson’s correlation coefficient; b, standardized coefficient; AFL, atrial flutter;
adjusted R 2 ¼ 0.28.
Figure 2 Receiver-operating characteristic curve for the flutter
cycle length to predict the risk of sinus node dysfunction requiring a pacemaker implantation.
Independent predictors of a pacemaker
implantation for sinus node dysfunction
In the univariate logistic regression analysis, a female gender, prior
cardiac surgery, structural heart disease, LVEF ,50%, atypical
flutter waves, and FCL of .273 ms emerged as predictors of
According to earlier studies,6,8 atrial remodelling can be defined as
electroanatomic and EP abnormalities of the atria characterized by
atrial dilatation, a lower atrial voltage, and conduction slowing, and
these alterations in the atria should result from a loss of functioning
atrial myocardium. Stiles et al. 8 reported that patients with AFL had
evidence of this atrial remodelling even though they were studied
remote from an arrhythmic event, and the authors proposed that
the atrial remodelling was responsible for the predisposition to the
development of AFL. As mentioned in the introduction, based on
the ‘circus movement theory,4 the FCL of CTI-dependent AFL
would be expected to be prolonged when the RA, including the
TA, is dilated or when the conduction velocity of the atria is
slower or both. As mentioned above, these atrial alterations
should be considered as a type of atrial remodelling. Accordingly,
it is hypothesized that the FCL reflects the extent of the atrial
remodelling. We importantly observed a positive correlation
between the FCL and atrial conduction time, which should
support this hypothesis. Furthermore, a following experiment
with sheep14 reinforced the inference as well. According to the
researchers, while the FCL became progressively shorter only in
the early period after the onset of the AFL, in the subsequent
period the FCL gradually prolonged along with slowing of the
atrial conduction velocity. The investigators of the reports on
the classification of AFL and AT15 explicitly stated that a conduction delay within the circuit could prolong the FCL. Indeed, a considerable number of patients recruited in the studies conducted by
Akar et al. 16 and Stevenson et al. 17 who had scars or incisions in
the RA possibly leading to a conduction delay presented with a
relatively longer FCL of the CTI-dependent AFL. Those reports
may also confirm the aforementioned concept that significant
remodelling of the atria prolongs the FCL.
Relationship between the atrial
remodelling and sinus node dysfunction
Stevenson et al. 17 also reported that five of eight patients with RA
scars demonstrated evidence of significant SND requiring a PMI.
Furthermore, several other studies have clearly shown that subjects with SND demonstrated diffuse atrial remodelling.5 – 8 It is
known that the pacemaker complex in humans extends widely
from the superior vena cava-RA junction to the inferior vena
Downloaded from by guest on October 24, 2016
20.13
Major findings
20.048
0.1
0.2
Left atrium diameter
Ventricular cycle length
during AFL
Discussion
385
Prediction of SND in patients with persistent AFL using the FCL
Table 4 Predictors of an implantation of a pacemaker for sinus node dysfunction in the univariate and multivariate
analyses (n 5 211)
Variable
Univariate
.............................................
OR (95% CI)
P value
Multivariate
..............................................
OR (95% CI)
P value
...............................................................................................................................................................................
Age .70 years
1.51 (0.99– 2.36)
0.06
Female
Duration of atrial flutter .3 months
1.97 (1.24– 3.09)
0.81 (0.39– 1.45)
0.0032
0.53
2.43 (1.32–4.62)
0.0046
1.50 (0.78–2.83)
0.21
2.10 (1.20–3.87)
0.012
5.34 (3.08–10.08)
,0.0001
Prior cardiac surgery
1.89 (1.15– 3.02)
0.009
Structural heart disease
1.71 (1.11– 2.65)
0.015
Hypertension
Diabetes mellitus
1.22 (0.79– 1.88)
1.02 (0.54– 1.73)
0.36
0.94
Antiarrhythmic agents or digitalis
1.06 (0.69– 1.64)
0.79
Left ventricular ejection fraction ,50%
Left atrial diameter .40 mm
1.61 (1.04– 2.48)
1.22 (0.77– 1.93)
0.031
0.39
Atypical flutter waves
2.73 (1.50– 4.89)
0.0007
Flutter cycle length .273 ms
Ventricular cycle length during AFL ,500 ms
5.15 (3.14– 9.00)
0.81 (0.33– 1.94)
,0.0001
0.61
OR, odds ratio; CI, confidence interval; AFL, atrial flutter.
Left ventricular ejection fraction
and sinus node dysfunction
Another independent predictor of SND requiring a PMI was a
decreased LVEF. This was confirmed by Sanders et al.,7 who demonstrated that patients with congestive heart failure and thereby a
higher likelihood of a lower LVEF showed significant sinus node
remodelling. The authors speculated that both chronic stretch and
neurohormonal activation might have played a role.
Previous cardiac surgery and sinus
node dysfunction
Sinus node dysfunction after open-heart surgery has repeatedly
been identified in previous studies,18,19 and a trend towards it
was also observed in the present analysis. The researchers proposed direct damage to the sinus node or sinus node blood
supply during cardiac surgery as the causal mechanism. We
observed a significant trend towards an association between atypical flutter waves and SND, and this may be causally related to a
previous open-heart surgery as well. This is because atypical
flutter waves are often identified in patients with prior cardiac
operations,15 in fact, about two-third of the patients with that
ECG finding had previously undergone such in our study.
Temporal recovery of the sinus
node function
Daoud et al. 2 showed a progressive temporal recovery of the
CSNRT among the patients who underwent ablation for persistent
AFL, which supports our finding that one-fourth of the patients
requiring temporary pacing for significant SND following termination of AF recovered their sinus node function within 2 weeks
after the CA. The proportion of an exteriorization of significant
SND after the termination of AFL in our study appeared to be
somewhat higher, possibly because an arrhythmogenic substrate
causing the AFL was present in the atrium of the patients with
SND as previously indicated.8,20,21
Ability of the corrected sinus node
recovery time to identify subjects
requiring a pacemaker implantation
In our study, the ability of the CSNRT to discriminate between
patients with and without clinically significant SND was somewhat
limited and even lower than that of the FCL, and in particular, the
sensitivity of the cut-off value was relatively low. This may be
because indeed EP testing including an overdrive suppression
test can accurately delineate fixed cardiac conduction defects but
its ability to identify SND taking the form of transient sinus
Downloaded from by guest on October 24, 2016
cava-RA junction along the long axis of the crista terminalis. Based
on this nature of the pacemaker complex, Sanders et al.6 suggested
that a significant derangement of the sinus node function would
require extensive loss of automatic pacemaker tissue and widespread impairment of the sinoatrial conduction. To put it the
other way around, it is inferred that a loss of normal tissue and
a scarring expanding all over the RA can result in significant
SND, which should be a reasonable explanation for the relationship between the atrial remodelling and SND that the aforementioned studies demonstrated. We in this study showed that the
conduction time of the atria was directly proportional to the
CSNRT among the patients with persistent AFL, which was consistent with the findings of those studies when considering the definition of the atrial remodelling provided in the start of the
discussion. Thus, it can safely be said that SND is significantly
associated with diffuse atrial remodelling. When this inference is
taken with the aforementioned concept that an increased FCL
reflects atrial remodelling, a significant rationale for our major
finding of a significant relationship between a prolonged FCL and
SND may be provided.
386
arrest and sinus bradycardia is limited.22,23 Furthermore, a previous
study reported that EP testing sometimes reveals unrelated rhythm
disturbances that can lead to syncope,23 which may help explain
why pacemakers were not implanted in some patients despite
their significantly longer CSNRT. In addition, as mentioned previously,24 an insufficient reproducibility of the CSNRT may also
be responsible for its limited ability to identify subjects needing a
pacemaker. These discussions might possibly lead to the notion
that the FCL more accurately reflects the sinus node function
than the CSNRT in each individual; however, our study recruited
a relatively specific population; patients with sustained isthmusdependent AFL, and therefore, predicting SND by using the FCL
cannot be applied to the general populations. Furthermore, in
the current consensus, an overdrive suppression test has been
recognized as the standard EP test to assess the sinus node function. Accordingly, the FCL cannot take the place of the CSNRT
for differentiating subjects with and without clinically significant
SND.
Limitations
Clinical implications
To the best of our knowledge, this is the first study of an association between the FCL and sinus node function. The FCL can be
assessed using the surface ECG as well. In light of our major
finding, therefore, assessing it prior to ablation can be helpful for
the risk stratification of SND requiring a PMI in patients with persistent AFL. Also, our results suggest that the FCL may yield some
important information regarding the atrial remodelling.
Acknowledgement
The authors acknowledge Mr. John Martin for his grammatical
assistance with this paper.
Conflict of interest: none declared.
References
1. Blomström-Lundqvist C, Scheinman MM, Aliot EM, Alpert JS, Calkins H, Camm AJ
et al. ACC/AHA/ESC guidelines for the management of patients with supraventricular arrhythmias –executive summary. a report of the American college of cardiology/American heart association task force on practice guidelines and the
European society of cardiology committee for practice guidelines (writing committee to develop guidelines for the management of patients with supraventricular arrhythmias) developed in collaboration with NASPE-Heart Rhythm Society.
J Am Coll Cardiol 2003;42:1493 –1531.
2. Daoud EG, Weiss R, Augostini RS, Kalbfleisch SJ, Schroeder J, Polsinelli G et al.
Remodeling of sinus node function after catheter ablation of right atrial flutter.
J Cardiovasc Electrophysiol 2002;13:20–4.
3. Dobrzynski H, Boyett MR, Anderson RH. New insights into pacemaker activity:
promoting understanding of sick sinus syndrome. Circulation 2007;115:1921 –32.
4. Frame LH, Page RL, Hoffman BF. Atrial reentry around an anatomic barrier with a
partially refractory excitable gap. A canine model of atrial flutter. Circ Res 1986;58:
495 –511.
5. Elvan A, Wylie K, Zipes DP. Pacing-induced chronic atrial fibrillation impairs sinus
node function in dogs. Electrophysiological remodeling. Circulation 1996;94:
2953 –60.
6. Sanders P, Morton JB, Kistler PM, Spence SJ, Davidson NC, Hussin A et al.
Electrophysiological and electroanatomic characterization of the atria in
sinus node disease: evidence of diffuse atrial remodeling. Circulation 2004;109:
1514 –22.
7. Sanders P, Kistler PM, Morton JB, Spence SJ, Kalman JM. Remodeling of sinus node
function in patients with congestive heart failure: reduction in sinus node reserve.
Circulation 2004;110:897–903.
8. Stiles MK, Wong CX, John B, Kuklik P, Brooks AG, Lau DH et al. Characterization
of atrial remodeling studied remote from episodes of typical atrial flutter. Am J
Cardiol 2010;106:528–34.
9. Sairaku A, Eno S, Hondo T, Teragawa H, Nakano Y, Matsuda K et al.
Head-to-head comparison of the cardio-ankle vascular index between patients
with acute coronary syndrome and stable angina pectoris. Hypertens Res 2010;
33:1162 –6.
10. Tada H, Oral H, Sticherling C, Chough SP, Baker RL, Wasmer K et al. Double
potentials along the ablation line as a guide to radiofrequency ablation of
typical atrial flutter. J Am Coll Cardiol 2001;38:750–5.
11. Morton JB, Sanders P, Deen V, Vohra JK, Kalman JM. Sensitivity and specificity of
concealed entrainment for the identification of a critical isthmus in the atrium:
relationship to rate, anatomic location and antidromic penetration. J Am Coll
Cardiol 2002;39:896–906.
12. Maury P, Duparc A, Hebrard A, El Bayomy M, Delay M. Prevalence of typical atrial
flutter with reentry circuit posterior to the superior vena cava: use of entrainment at the atrial roof. Europace 2008;10:190 –6.
13. Miyanaga S, Yamane T, Date T, Tokuda M, Aramaki Y, Inada K et al. Impact of pulmonary vein isolation on the autonomic modulation in patients with paroxysmal
atrial fibrillation and prolonged sinus pauses. Europace 2009;11:576 –81.
14. Morton JB, Byrne MJ, Power JM, Raman J, Kalman JM. Electrical remodeling
of the atrium in an anatomic model of atrial flutter: relationship between
substrate and triggers for conversion to atrial fibrillation. Circulation 2002;105:
258 –64.
15. Saoudi N, Cosı́o F, Waldo A, Chen SA, Iesaka Y, Lesh M et al. A classification of
atrial flutter and regular atrial tachycardia according to electrophysiological mechanisms and anatomical bases; a Statement from a Joint Expert Group from The
Working Group of Arrhythmias of the European Society of Cardiology and the
North American Society of Pacing and Electrophysiology. Eur Heart J 2001;22:
1162 –82.
16. Akar JG, Kok LC, Haines DE, DiMarco JP, Mounsey JP. Coexistence of type I atrial
flutter and intra-atrial re-entrant tachycardia in patients with surgically corrected
congenital heart disease. J Am Coll Cardiol 2001;38:377 –84.
17. Stevenson IH, Kistler PM, Spence SJ, Vohra JK, Sparks PB, Morton JB et al. Scarrelated right atrial macroreentrant tachycardia in patients without prior atrial
surgery: electroanatomic characterization and ablation outcome. Heart Rhythm
2005;2:594 –601.
18. Koplan BA, Stevenson WG, Epstein LM, Aranki SF, Maisel WH. Development and
validation of a simple risk score to predict the need for permanent pacing after
cardiac valve surgery. J Am Coll Cardiol 2003;41:795 –801.
19. Onalan O, Crystal A, Lashevsky I, Khalameizer V, Lau C, Goldman B et al. Determinants of pacemaker dependency after coronary and/or mitral or aortic valve
surgery with long-term follow-up. Am J Cardiol 2008;101:203 –8.
20. Anand N, McCrindle BW, Chiu CC, Hamilton RM, Kirsh JA, Stephenson EA et al.
Chronotropic incompetence in young patients with late postoperative atrial
flutter: a case-control study. Eur Heart J 2006;27:2069 –73.
Downloaded from by guest on October 24, 2016
As indicated by a previous investigation,25 the FCL is affected by
the autonomic tone. Therefore, not conducting an autonomic
blockade was a major limitation. The anatomical characteristics
of the RA could yield significant information on the atrial remodelling,26 we, however, did not routinely assess them. Also, there are
some variations in the arrhythmic circuit even in typical AFL,12
which may have influenced the FCL. We enrolled patients with
‘persistent’ AFL; however, the duration of persistent or chronic
AFL may be difficult to define accurately due to the misinterpretation of the symptoms. In our study, patients were not included if
they had a previous PMI or an inability to assess the sinus node
function due to not returning to sinus rhythm. This possibly led
to underestimating the prevalence of the coexistence of sustained
AFL and clinically significant SND. Moreover, not assessing the
long-term outcome regarding the sinus node function among the
patients was also an important limitation. Finally, this was a retrospective study and the data analysed were from a non-selected
population. Therefore, it is possible that several predictors of
SND requiring a PMI determined in our study would not be clinically significant enough to be directly applied to clinical practice.
Thus, it is desirable to conduct a prospective study including
selected populations in the future.
A. Sairaku et al.
Prediction of SND in patients with persistent AFL using the FCL
21. de Groot NM, Schalij MJ. The relationship between sinus node dysfunction,
bradycardia-mediated atrial remodeling, and post-operative atrial flutter in
patients with congenital heart defects. Eur Heart J 2006;27:2036 –7.
22. Taneja T, Mahnert BW, Passman R, Goldberger J, Kadish A. Effects of sex and age
on electrocardiographic and cardiac electrophysiological properties in adults.
Pacing Clin Electrophysiol 2001;24:16– 21.
23. Fujimura O, Yee R, Klein GJ, Sharma AD, Boahene KA. The diagnostic sensitivity
of electrophysiologic testing in patients with syncope caused by transient bradycardia. N Engl J Med 1989;321:1703 –7.
387
24. de Marneffe M, Jacobs P, Englert M. Reproducibility of electrophysiologic parameters of extrinsic sinus node function in patients with and without sick sinus
syndrome. Pacing Clin Electrophysiol 1986;9:482 –9.
25. Stambler BS, Ellenbogen KA. Elucidating the mechanisms of atrial flutter cycle length
variability using power spectral analysis techniques. Circulation 1996;94:2515 – 25.
26. Da Costa A, Mourot S, Roméyer-Bouchard C, Thévenin J, Samuel B, Kihel A et al.
Anatomic and electrophysiological differences between chronic and paroxysmal
forms of common atrial flutter and comparison with controls. Pacing Clin Electrophysiol 2004;27:1202 – 11.
Downloaded from by guest on October 24, 2016