CLINICAL RESEARCH
Europace (2016) 18, 274–280
doi:10.1093/europace/euv097
Cardiac electrophysiology
Electroanatomic characteristics of the mitral
isthmus associated with successful mitral isthmus
ablation
Decebal Gabriel Latcu 1*, Fabien Squara2,3, Youssef Massaad 1, Sok-Sithikun Bun 1,
Nadir Saoudi 1, and Francis E. Marchlinski 2
1
Cardiologie, Centre Hospitalier Princesse Grace, Avenue Pasteur, Monaco 98000, Monaco; 2Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104, USA; and
CHU Pasteur, 30 Avenue de la Voie Romaine, Nice 06000, France
3
Received 29 December 2014; accepted after revision 17 March 2015; online publish-ahead-of-print 23 December 2015
Aims
The success of mitral isthmus (MI) ablation has been related to CT scan defined MI anatomy. We sought to correlate
electroanatomical MI characteristics with MI ablation success in patients with perimitral flutter (PMF).
.....................................................................................................................................................................................
Methods
In 53 consecutive patients (46 males, 61 + 10 years) with PMF, MI was ablated with endocardial + coronary sinus (CS)
linear radiofrequency (RF) ablation lesion. Acute (termination of PMF during ablation) and long-term procedural success
and results
were studied. Mitral isthmus characteristics (thickness—minimal endocardial to CS distance, length, maximal MI bipolar
voltage), as well as MI ablation line length and width, RF duration, and delivered energy were analysed. In 43 of the 53
patients (81%), acute success was observed. This was more frequently achieved in patients with thinner MI
(2.4 + 3.1 vs. 7 + 3.2 mm; P ¼ 0.0009). Mitral isthmus thickness predicted ablation failure with a ROC area of 0.84.
The best threshold to predict MI ablation failure was 8.3 mm with a sensitivity of 67% and a specificity of 97%. Left
atrial size was of greater importance in failed cases (2D echo surface: 24.1 + 2.5 vs. 32.5 + 6.9 cm2, P ¼ 0.005;
electroanatomic volume: 124 + 32 vs. 165 + 23 mL, P ¼ 0.02). None of the other electroanatomical characteristics
were associated with outcome. After a mean follow-up of 28 + 15 months, 21 patients (39%) had atrial fibrillation
(AF) or atypical flutter (PMF recurrence in four).
.....................................................................................................................................................................................
Conclusion
Smaller MI thickness is associated with acute success in PMF ablation. Mitral isthmus electroanatomical characteristics
might be used for decision-making on strategy during persistent AF ablation and for selecting the best location for interrupting PMF.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Perimitral flutter † Catheter ablation † Electroanatomic mapping
Introduction
Perimitral flutter (PMF) is a challenging arrhythmia, most often occurring after ablation of atrial fibrillation (AF).1 Strategies of persistent
AF ablation include the creation of a line between the inferior
aspect of the left inferior pulmonary vein (LIPV) –left atrium (LA)
junction and the lateral aspect of the mitral valve (MV). Creation of
this mitral isthmus (MI) line2 is accompanied by a high percentage
of recovery of conduction which is arrhythmogenic and actually
favours the occurrence of PMF.3,4
The most widespread ablation strategy for PMF is MI ablation
with endpoints of PMF termination and block across the line.2
These endpoints are difficult to obtain despite the use of epicardial
ablation inside the coronary sinus (CS); pitfalls in diagnosis of MI
block render this task even more difficult.5 Reported success
rates are widely distributed (56 – 96% for MI block2,3,6 – 9 and 88 –
100% for PMF termination1,8). Alternative strategies have been proposed including targeting other anatomic locations to interrupt the
PMF.10 – 13 Injury of the circumflex artery seems frequent with MI
ablation.14
Although several studies have described the anatomy of the MI in
pathologic specimens,15,16 and by CT-scan imaging,3,17 none has examined the impact of the isthmus anatomy assessed with an electroanatomic mapping (EAM) system in the electrophysiology (EP) lab.
* Corresponding author. Tel: +377 97 98 97 71; fax: +377 97 98 97 32. E-mail address: [email protected]
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015. For permissions please email: [email protected].
275
Electroanatomic characteristics of the mitral isthmus
What’s new?
† This is the first study to correlate electroanatomical MI characteristics with MI ablation success in patients with PMF.
† We found that a smaller MI thickness (,8.3 mm) is associated
with acute success in PMF ablation.
† This results have the potential to strongly impact our practice,
as MI electroanatomical characteristics might be used for
decision-making on strategy during persistent AF ablation
and for selecting the best location for interrupting PMF.
We aimed to identify MI morphology and ablation line characteristics, as assessed in the EP lab, which may influence the acute efficacy
of linear ablation in PMF patients.
Methods
Data of patients who underwent radiofrequency (RF) ablation for
PMF at Centre Hospitalier Princesse Grace (CHPG; Monaco) and
at the Hospital of the University of Pennsylvania (HUP) were
reviewed. The study was approved by the institutional committee
on human research. According to institutional guidelines all patients
gave written informed consent for the invasive procedures. Antiarrhythmic drugs (except amiodarone) were withdrawn at least five
half-lives before the procedures. All patients had pre-procedural
echocardiograhic assessment of the LA (antero-posterior diameter
in the parasternal long axis view, 2D surface in the apical fourchamber view) and of left ventricle (LV) ejection fraction by the Simpson’s biplane method.
Ablation procedure
The approach for LA arrhythmia ablation in each institution has been
described elsewhere.18,19 In brief, we performed CS catheterization
with decapolar diagnostic catheter, double transeptal puncture, systemic anticoagulation with heparin (target-activated clotting time
.350 s), and EAM of the LA with the Carto system (BiosenseWebster Inc., Diamond Bar, CA, USA; BW). Mapping and ablation
catheters were inserted transseptally via a non-steerable (Fast-Cath
SL1, St. Jude Medical, Minnetonka, MN, USA; SJM) or steerable (Agilis,
SJM or V-Cas Deflect, Stereotaxis) sheath. A circular mapping catheter was used to assess pulmonary vein (PV) potentials.
Mapping and ablation
Detailed activation mapping of the LA was performed. Left atrium
shell for anatomic definition was done with either standard electroanatomical (EA; Carto XP) or adjusted fast anatomical techniques
(FAM; Carto 3). Perimitral circuit was confirmed by LA and CS activation and entrainment mapping. In all procedures, the operator systematically delivered point-by-point ablation lesions for at least 30 s
to create a contiguous MI line between the inferior margin of the left
PV encircling line and the lateral aspect of the MV (at 3– 4 o’clock).
Catheter –tissue contact was optimized before each RF delivery,
using catheter motion on fluoroscopy, near-field electrogram
(EGM) stability, impedance drop during RF delivery and morphologic
EGM changes suggestive of lesion creation.18 Whenever available,
contact force was used. If a significant lesion (based on local electrogram modification) was not obtained, reablation at the same site with
further optimization of contact (steerable sheath) and energy increase was performed. Irrigated RF was delivered with a Stockert
70 generator (BW), a 428C limit and 30 –40 W for the endocardial
part of the line. Baseline irrigation flow was 17– 30 mL/min. All
patients had antral PV encircling lines with PV isolation.
A separate electroanatomic map was created for the CS focusing
on information in the region facing the endocardial line.
Acute success was defined as termination of PMF with sinus
rhythm (SR) resumption or its change to another stable atrial tachycardia (AT) during MI ablation (without atrial extrasystole) shown
by activation or entrainment mapping not to be PMF. An ablationinduced transformation to another stable AT was presumed if an
abrupt and sustained change in cycle length, intra-cardiac activation
pattern and surface ECG occurred. Minor (e.g. ,10%) prolongation
of cycle length alone was not accompanied by changes of the intracardiac activation pattern and/or surface ECG atrial wave morphology
was not considered as an acute success. If the PMF was unchanged
after MI ablation, failure was considered.
Conduction block was assessed by the activation sequence along
the CS during left atrial appendage or lateral MV pacing.20 Complete
block was defined by proximal-to-distal activation on the septal side
of the line during pacing laterally to the line. In case of ongoing PMF or
in absence of a block after the completion of the endocardial MI line,
ablation lesions were delivered epicardially, inside the CS, facing the
endocardial line. The power setting was set at 20 –25 W. Ablation
was continued up to complete abatement of the local potentials in
this region, and block reassessed.
Additional ablation was performed, if necessary, for PV disconnection. Associated AT ablation was performed at the critical isthmus or
focal origin, up to SR resumption. Mitral isthmus block was reconfirmed
before catheter withdrawal (with a minimum 30 min waiting time).
Mitral isthmus and ablation line
measurements
Mapping data were reviewed offline retrospectively by two observers
experienced with the Carto system and blinded to the procedural
result. Left atrium and CS maps were simultaneously displayed. Local
resolution in the MI area for the LA FAM and for the CS FAM maps
were optimized to the maximal value (labelled ‘20’ in the Carto software). The ‘distance measurement’ and ‘design line’ tools were used.
For cases with EA maps, care was taken to measure the distance to
an acquired point (e.g. in a floating manner) inside the CS in order to
avoid minimization bias due to interpolation. To minimize errors,
each measurement was repeatedly performed and 3D manipulation
used to ensure correct placement of the distance tool. The following
MI and ablation line characteristics were measured
(1) MI thickness is defined as the minimal distance between endocardial and CS high-resolution maps (or from the LA map to the
closest acquired point within the CS in case of EAM) at the MI
level (Figure 1);
(2) MI length is defined as the minimal distance between the inferior
margin of the LIPV –LA junction and the lateral aspect of the MV
(Figure 2A);
276
D.G. Latcu et al.
Figure 1 Mitral valve (blue ring) views (caudal LAO) of the LA and CS maps. Mitral isthmus thickness is measured as the minimal distance between
endocardial LA and CS high-resolution maps at the MI level. (A) Example of overlapping LA and CS maps (0 mm thickness of the MI). (B) Example of a
‘thick’ MI (8.6 mm measured with the ‘Distance measurement’ tool).
Figure 2 (A) Mitral isthmus length is minimal distance between the inferior margin of the LIPV – LA junction) and the lateral aspect of the MV. (B)
The MI ablation ‘line’ curvilinear length, measured with the ‘Design Line’ tool, following the centre of the lesions from the venous to the mitral aspect
of the MI line. (C) The MI ablation line width, measured between the border of the most lateral and the most septal lesions.
(3) maximal MI bipolar EGM voltage during PMF;
(4) MI ablation line curvilinear length, as measured with the ‘Design
line’ tool (Figure 2B);
(5) actual MI ablation line width (Figure 2C), which we felt would
reflect the difficulties in lesion creation;
(6) RF application duration and energy endocardially and epicardially
(within the CS).
variables were compared with the t-test or the Mann– Whitney–
Wilcoxon test (in case of non-normality). Nominal variables were
compared using x2-test or Fisher’s exact, as appropriate. P , 0.05
was considered significant. The predictive values of different thresholds of MI characteristics for ablation success were assessed using
sensitivity, specificity, and the receiving operator characteristics
(ROC) curve analysis.
Follow-up
Follow-up for clinical and asymptomatic recurrences was performed
in a ‘real-life’ setting by regular visits (including ECG) to the treating
cardiologist with repeated Holter monitoring in all cases (every 3
months). Any recurring, sustained, symptomatic AF or flutter was
considered for a repeat procedure.
Statistical analysis
Continuous variables are expressed as mean + SD. Median values
and 95% CI are given in cases of non-normal distribution. Numerical
Results
Study population
Fifty-three patients (CHPG: 30, HUP: 23; males 87%; 60.9 + 10.1
years) undergoing RF ablation for PMF (2007–2014) were retrospectively included. An endocardial MI line was performed in all
patients. Patient characteristics are summarized in Table 1.
A majority of patients had a history of AF ablation (75%) and in 15
patients (28%) a MI line had previously been created (as a step in a
277
Electroanatomic characteristics of the mitral isthmus
Table 1 Patient characteristics
Patient characteristics
Overall (n 5 53)
Successful MI line
81% (n 5 43)
Failure MI line
19% (n 5 10)
P-value
...............................................................................................................................................................................
Sex ratio (males)
46 (87%)
37 (86%)
9 (90%)
1
Age
60.9 + 10.1
60.5 + 10.3
62.3 + 9.5
0.51
Hypertension
Smoking History
22
11 (21%)
21 (49%)
9 (21%)
1 (10%)
2(20%)
0.03
1
Diabetes
5 (9.4%)
5 (11.6%)
0 (0%)
0.57
Obesity
Dyslipidemia
9 (17%)
16 (30%)
7 (16%)
13 (30%)
2 (20%)
3 (30%)
1
1
Coronary heart disease
5 (9.4%)
5 (11.6%)
0 (0%)
0.57
Hypertensive cardiomyopathy
Valvular heart disease
6 (11.6%)
5 (9.4%)
5 (11.6%)
4 (9.3%)
1 (10%)
1 (10%)
1
1
History of AF ablation
40 (75%)
32 (74%)
8 (80%)
1
History of prior left isthmus ablation
LA antero-posterior diameter
15 (28%)
52.7 + 7.2 mm
12 (27.9%)
51.8 + 6.4 mm
3 (30%)
55 + 9.3
1
0.45
LA surface
27.8 + 6.4 cm2
24.1 + 2.5 cm2
32.5 + 6.9 cm2
0.005
LV ejection fraction
55 + 14%
56 + 13%
50 + 17%
0.28
persistent AF ablation strategy in all). Left atrium was larger (as evaluated by the ultrasound surface) in the unsuccessful PMF ablation
group. There were more hypertensive patients in the successful
PMF ablation group (see Table 1).
Ablation procedure and acute success
General anaesthesia was used in 15 patients (28%), with jet ventilation in three. Manual irrigated catheters [Thermocool (n ¼ 32) or
Smartouch (n ¼ 6), BW] were used for mapping and ablation in 38
(72%) patients, whereas the remaining 15 (28%) cases were performed with magnetic navigation (Thermocool RMT, BW). Steerable
sheaths were used in 30 cases (57%): Agilis (SJM) for manual cases
(26/38) or V-Cas Deflect (Stereotaxis) for magnetic cases (4/15).
The use of steerable sheaths improved ablation result (90% vs.
68% for PMF termination, P ¼ 0.04). No difference was noticed
between manually guided catheters and magnetic navigation. All
the cases performed with the steerable robotic sheath (V-Cas
Deflect; n ¼ 4) were successful.
Perimitral flutter (mean CL 265 + 55 ms) was the unique AT in
16 patients (30%) but the majority of patients had several AT’s
(up to 5, median: two different AT/patient). The PMF was clockwise
in 46%, counter-clockwise in 46% and alternated both rotations
in 8%.
In 40 of the 53 patients (75%) epicardial ablation in the distal CS
was required. Acute success rate of the MI ablation was 81% (43/
53 patients); in only one failed case an alternative line (anterior)
was successfully performed. Despite PMF interruption, MI block
was obtained in only 57% of the study population.
Sinus rhythm resumed directly in 23 of the 43 successful cases
(53%). In the remaining 20 cases, PMF transformed into another
AT: left atrial in 18 (42%) and right atrial (typical flutter) in 2 (5%). Successful ablation of all AT with SR restoration was obtained in 70% of
the study population.
Total RF delivery time per procedure was 29 + 14 min. Mean RF
delivery time for the endocardial MI line was 15 + 12 min, whereas
RF delivery time within the CS was 4 + 3 min; although there was
no significant difference in RF delivery time between successful or
failed cases, there was a strong trend towards a larger amount of
energy delivered endocardially (but not epicardially) in failed cases
(Table 2).
Mean procedure duration was 261 + 67 min (95% CI 236–
286 min); there was no difference in procedure duration between
successful/failed cases. Procedural steps of endocardial MI line creation took 35 + 23 min, and epicardial additional ablation step
12 + 9 min. Fluoroscopy exposure time was 26 + 27 min. Only
one procedural complication occurred in this series, a small pericardial effusion that was managed conservatively.
Electroanatomical data
Detailed EA data are presented in Table 2. Increased LA size was confirmed by LA map volume (129 + 34 mL) that was significantly larger
in failed cases. Mitral isthmus was significantly thicker in failure group
(7 + 3.2 vs. 2.4 + 3.1 mm, P ¼ 0.0009). A ‘receiving operator characteristics’ curve was made for MI thickness (Figure 3). Prediction
value of the MI thickness was very good, with an area under curve
(AUC) of 0.84 (95% CI 0.67– 1). The best threshold which predicts
success/failure of PMF ablation was 8.3 mm, which correctly classified
91% of cases (sensitivity of 67% and specificity of 97%).
No other EA characteristic was predictive of PMF ablation failure
but there was a very strong trend towards a wider endocardial ablation line in failed cases (26.6 + 20 vs. 13.3 + 7.3 mm, P ¼ 0.05).
Outcome during follow-up
After a follow-up of 28 + 15 months after the index procedure (3–
48, median 33), an arrhythmia recurrence was documented in 21
patients (39%) after 16 + 16 months (median 7, range 2 weeks–48
278
D.G. Latcu et al.
Table 2 Electroanatomic characteristics
Overall (n 5 53)
Electroanatomic characteristics
Successful endo/epi
MI line 81% (n 5 43)
Failure endo/epi
MI line 19% (n 5 10)
P-value
...............................................................................................................................................................................
129 + 34
124 + 32
165 + 23
3.3 + 3.6
2.4 + 3.1
7 + 3.2
Maximal MI voltage during PMF (mV)
MI length (mm)
1.47 + 1.2
31.5 + 7.1
1.61 + 1.3
31.2 + 6.8
Ablation line length (mm)
43.4 + 13.7
LA map volume (ml)
MI thickness (mm)
Ablation line width (mm)
Endocardial MI line delivered energy (Ws)
Epicardial delivered energy (Ws)
26.6 + 20
16 561 + 6054
0.05
0.06
1656 + 3028
1855 + 3788
1324 + 1209
0.47
8.3 mm
Sensitivity
0.50
MI thickness
Area under ROC curve = 0.84
Reference
0.00
0.25
0.50
1-Specificity
0.16
13.3 + 7.3
8865 + 5981
0.75
0.00
49.1 + 13.2
0.19
0.76
16 + 11.9
11 233 + 6832
1.00
0.25
42 + 13.7
1.01 + 0.72
32.6 + 8.7
0.02
0.0009
0.75
1.00
Figure 3 Receiving operator characteristics curve of MI thickness
for ablation result predictability. The area under the ROC curve is
0.84 (95% CI 0.67– 1). The best threshold to predict success/
failure of PMF ablation was 8.3 mm, which correctly classified
91% of cases (sensitivity of 67% and specificity of 97%).
months). Recurrences were left AT other than PMF (12 patients,
53%), AF in 5 (24%) and PMF in 4 (19%). The four PMF cases recurred
at 3, 5, 36 and 48 months after the index procedure. Pulmonary vein
isolation was confirmed in all at the moment of the redo procedure.
Initial successful ablation of all AT and SR restoration was not predictive of recurrences nor of recurrence type, but recurring PMF
tended to be more frequent in failed PMF ablation cases (50% vs.
11% for the initially successful PMF ablation, P ¼ 0.12). Block
across the MI line had no impact on PMF recurrence.
Discussion
The main finding of this study is that MI thickness, as measured on the
EA mapping system in the EP lab, can accurately predict acute success
of PMF ablation.
Mitral isthmus thickness
Thickness of the MI has previously been assessed in the literature.
Histological analysis of autopsy specimens showed highly variable
myocardial thickness at the MI level (1.2 –4.4 mm).15,16 As measured
with the electroanatomic system, the MI thickness is the distance
between the endocardium and the CS inner surface facing the LA epicardium. It encompasses, on top of the myocardial thickness, the
adipose tissue overlaying the CS along the MV, sometimes with circumflex artery branches in between.16 Because of the possible position of the CS at some distance from the atrial wall, with the
electroanatomic system we may not be measuring myocardial thickness, but a real distance between the CS and the LA wall. Anatomical
studies show that in the vast majority of the cases, the CS runs along
the postero-inferior wall of the left atrium. Thus, a study of 61 excised
human cadaveric hearts21 showed a CS to MA distance up to 19 mm
(mean 9.7 mm) and, in all cases, the CS was superiorly located in
respect to the MA (range 1 –19 mm). In another CT-scan study of
the CS anatomy,22 an ever bigger CS to MA distance was measured
(14.2–16 mm).
In the literature, challenging cases of MI ablation were not reported
to be associated with a ‘thicker’ MI; rather a heat-sink hypothesis due
to blood flow within the CS11 and circumflex artery branches3 has
been proposed.
The CT-scan measures of MI have also been published.3,17 The
isthmus depth assessed by CT-scan has been correlated with
failure to achieve complete isthmus block.3 It is worth noting that
MI depth in these studies does not correspond to the true myocardial thickness but to the distance from the straight LIPV – MV line to
the deepest point of the endocardial isthmus. These authors
defined depth as a concept which suggests concave shape and a
lack of flatness rather than myocardial thickness. Even if myocardial
thickness itself measured on the CT-scan was not different
between groups, the interposing circumflex artery between the
MI and the CS predicted ablation failure. Surprisingly, this did not
translate into a difference in the minimum distance from the CS
to the endocardium between the two groups (3.2 vs. 3.7 mm,
P ¼ 0.27); this measure is the most appropriate comparator of
the MI thickness of the electroanatomy (mean value in our study
3.3 + 3.6 mm).
A visual assessment may also be used as a fair and rapid method for
PMF ablation success prediction. It is remarkable that in 23 cases
(43%), LA and CS electroanatomic maps overlapped in the MI
region (which translated into a ‘0’ mm MI thickness). Ablation was
successful in all but one.
279
Electroanatomic characteristics of the mitral isthmus
Mitral isthmus length
The MI length, as measured by electroanatomy in this study, is similar
to that measured on human hearts at autopsy (34–36 mm).15,16 In AF
patients, MI length was reported to be higher.17 Only one study
showed similar MI length between the electroanatomic reconstruction and the CT-scan.23 One electroanatomic study of the MI length
showed higher values (37 vs. 26 mm, P ¼ 0.008) in patients who
needed additional CS ablation to achieve MI block, and in patients
with very long MI (56 + 8 mm) block could not be achieved.7
In our study, a wider and, to a lesser extent, a longer ablation line
tends to be correlated with ablation failure. The ablation line may be
longer than the MI length not only due to concave/pouched shape
of the MI in the majority of cases17 but also due to local anatomical
anfractuosities/crevices16 at the level of the MI. The existence of
a pouch at the isthmus (as visualized by CT-scan) was reported
to be more frequent in patients with incomplete block (40% vs. 9%,
P ¼ 0.01).3
Outcome with ablation
In our series, ablation inside the CS was performed in 75% of the
cases. It is widely documented that a high percentage (60 –
74%)2,7,8,14 of patients need epicardial ablation inside the CS in
order to obtain MI block. No solid data exist about the factors
which determine the need for CS ablation, but a larger CS predicted
the need of epicardial RF delivery in one study.7
In the group with successful ablation, PMF terminated with direct
SR resumption in 53% of cases and by transformation into another AT
in 47%. Acute success with direct SR resumption is variable in the literature in case of PMF (26–55%).10 In our series, PMF acutely terminated in 81% (with simultaneous isthmus block in 57%); series
addressing specifically PMF are sparse, but reported success rate of
MI ablation (e.g. block creation) varies widely (56%– 92%).2,3,6,7,20
Long-term efficacy is the Achilles’ heel of PMF ablation. Recent
series3 report 39% recurrence of PMF, a mean of two procedures/
patient being needed in order to obtain a mid-term 84% efficacy
which is far more difficult than for cavotricuspid isthmus. Despite
acute bidirectional MI conduction block, recovery of conduction is
common (up to 73% of cases) and may lead to AT recurrence.4 A randomized study8 even showed that MI block had little impact on arrhythmia recurrence in patients presenting PMF after ablation for
longstanding persistent AF, alternative strategies of targeting venous
and non-venous triggers of arrhythmia being more efficient. Our
follow-up data show similar results; MI block does not impact longterm recurrence and overall arrhythmia recurrence rate (39%). In
our study, most recurrences were AF and non-PMF AT; only 19%
of recurrences were PMF (7.5% of the entire study population).
Clinical implications
The most important clinical implication of this study is that the simple
analysis of the CS –LA distance on electroanatomy helps in decisionmaking during persistent AF ablation, if the operator considers targeting the MI as part of the ablation procedure. This may be
coupled with pre-procedural imaging that documents an interposed
circumflex artery between the CS and the MI and also predicts lower
chances of a successful MI ablation.3 Our data suggests that a MI thickness .8.3 mm should discourage operators from performing the MI
line and make one consider a more anterior approach in patients with
documented PMF.
Limitations
The main limitation of our study is its retrospective design. Nevertheless, its main result (prediction value of the MI thickness) could not be
impacted by the retrospective nature of the assessment of this parameter. On the contrary, one may consider that operators, not
knowing the potential of this measure, performed the procedures
in a ‘blinded’ manner.
Follow-up was based on regular visits with Holter monitoring;
thus, recurrence rate might have been underestimated compared
with intermittent loop recorders.
The different technologies used in this series may be considered as
a limitation for data interpretation. No data exists on differences
between these approaches when ablating PMF, but this was beyond
our aim which was limited to identifying the MI morphology and ablation line characteristics influencing the acute efficacy of linear ablation of PMF.
Another possible limitation is the use of the PMF termination as the
acute success criterion instead of MI block creation. We acknowledge that the current guidelines recommend complete bidirectional
block across the linear lesion as the endpoint of MI ablation. Retrospective analysis of procedure reports and intracardiac recordings
provided reliable assessment of MI block in only 66% of the study
population. In absence of SR resumption at the moment of PMF termination, MI block could not be immediately assessed. Most importantly, pitfalls in the assessment of the MI linear conduction block have
been reported in 20% of the cases, thus rendering this diagnosis less
reliable than mere PMF termination.5 Finally, a randomized study8
showed that MI block creation has limited impact on arrhythmia recurrence and selected operators at our institutions may have not
required this endpoint. For the 35 patients in whom MI block was reliably assessed, it was obtained in 20 patients (57%) and there was a
very strong trend to a ‘thinner’ MI in these patients (2.9 + 3.1 vs.
5.3 + 4.1, P ¼ 0.05).
Patient movements and respiration artefacts on acquired volumes
are one important limitation during EAM. Use of intravenous sedation, general anaesthesia, jet ventilation, and respiratory compensation, whenever available, should have limited this bias. Nevertheless,
this limitation is systematically present and the results may thus be
extrapolated to everyday use in clinical practice. These artefacts, together with heart movements, explain overlapping of LA and CS electroanatomic maps in the MI region in a significant number of cases
(which translated into a ‘0’ mm MI thickness); we consider that in
these cases, MI is very ‘thin’ and this information is clinically valuable.
Manual measurements may also have impacted results. Observers
used 3D manipulation of the maps extensively in order to establish
the most appropriate boundaries for the distance callipers, and agreement between the two observers was sought before validating any
value (rendering intra- and inter-observer variability inappropriate).
Conclusion
The MI thickness measured by electroanatomy is predictive of acute
success of PMF ablation with a value .8.3 mm predicting ablation
failure with 91% accuracy. Mitral isthmus EA thickness might be
280
used to optimize the ablation strategy of persistent AF for identifying
alternative (e.g. anterior) locations of mitral annular line.
Acknowledgements
Preliminary results of this study have been presented at Cardiostim
2012 (Europace 2012;14(S1):136P–32).24
Conflict of interest: none declared.
Funding
F.S. has received a research grant from the Fédération Française de Cardiologie, Sorin Group, and Endosense.
References
1. Chugh A, Oral H, Lemola K, Hall B, Cheung P, Good E et al. Prevalence, mechanisms,
and clinical significance of macroreentrant atrial tachycardia during and following left
atrial ablation for atrial fibrillation. Heart Rhythm 2005;2:464 –71.
2. Jais P, Hocini M, Hsu LF, Sanders P, Scavee C, Weerasooriya R et al. Technique and
results of linear ablation at the mitral isthmus. Circulation 2004;110:2996 –3002.
3. Yokokawa M, Sundaram B, Garg A, Stojanovska J, Oral H, Morady F et al. Impact
of mitral isthmus anatomy on the likelihood of achieving linear block in patients
undergoing catheter ablation of persistent atrial fibrillation. Heart Rhythm 2011;8:
1404 –10.
4. Sawhney N, Anand K, Robertson CE, Wurdeman T, Anousheh R, Feld GK. Recovery
of mitral isthmus conduction leads to the development of macro-reentrant tachycardia after left atrial linear ablation for atrial fibrillation. Circ Arrhythm Electrophysiol
2011;4:832 –7.
5. Shah AJ PP, Miyazaki S, Liu X, Roten L, Derval N, Jadidi AS et al. Prevalence and types
of pitfall in the assessment of mitral isthmus linear conduction block. Circ Arrhythm
Electrophysiol 2012;5:957 –67.
6. Ernst S, Schluter M, Ouyang F, Khanedani A, Cappato R, Hebe J et al. Modification of
the substrate for maintenance of idiopathic human atrial fibrillation: efficacy of radiofrequency ablation using nonfluoroscopic catheter guidance. Circulation 1999;100:
2085 –92.
7. Wong KC, Jones M, Sadarmin PP, De Bono J, Qureshi N, Rajappan K et al. Larger coronary sinus diameter predicts the need for epicardial delivery during mitral isthmus
ablation. Europace 2011;13:555 –61.
8. Bai R, Di Biase L, Mohanty P, Dello Russo A, Casella M, Pelargonio G et al. Ablation of
perimitral flutter following catheter ablation of atrial fibrillation: impact on outcomes
from a randomized study (propose). J Cardiovasc Electrophysiol 2012;23:137 –44.
9. Ammar S, Luik A, Hessling G, Bruhm A, Reents T, Semmler V et al. Ablation of perimitral flutter: acute and long-term success of the modified anterior line. Europace
2015;17:447 –52.
D.G. Latcu et al.
10. Tzeis S, Luik A, Jilek C, Schmitt C, Estner HL, Wu J et al. The modified anterior line: an
alternative linear lesion in perimitral flutter. J Cardiovasc Electrophysiol 2010;21:
665 –70.
11. D’Avila A, Thiagalingam A, Foley L, Fox M, Ruskin JN, Reddy VY. Temporary occlusion of the great cardiac vein and coronary sinus to facilitate radiofrequency catheter
ablation of the mitral isthmus. J Cardiovasc Electrophysiol 2008;19:645 –50.
12. Baez-Escudero JL, Morales PF, Dave AS, Sasaridis CM, Kim YH, Okishige K et al.
Ethanol infusion in the vein of marshall facilitates mitral isthmus ablation. Heart
Rhythm 2012;9:1207 –15.
13. Berruezo A, Bisbal F, Fernandez-Armenta J, Calvo N, Angel Cabrera J,
Sanchez-Quintana D et al. Transthoracic epicardial ablation of mitral isthmus for
the treatment of recurrent perimitral flutter. Heart Rhythm 2013;11:26–33.
14. Wong KC, Lim C, Sadarmin PP, Jones M, Qureshi N, De Bono J et al. High incidence of
acute sub-clinical circumflex artery ‘injury’ following mitral isthmus ablation. Eur
Heart J 2011;32:1881 –90.
15. Becker AE. Left atrial isthmus: anatomic aspects relevant for linear catheter ablation
procedures in humans. J Cardiovasc Electrophysiol 2004;15:809 –12.
16. Wittkampf FH, van Oosterhout MF, Loh P, Derksen R, Vonken EJ, Slootweg PJ et al.
Where to draw the mitral isthmus line in catheter ablation of atrial fibrillation: histological analysis. Eur Heart J 2005;26:689 –95.
17. Chiang SJ, Tsao HM, Wu MH, Tai CT, Chang SL, Wongcharoen W et al. Anatomic
characteristics of the left atrial isthmus in patients with atrial fibrillation: lessons
from computed tomographic images. J Cardiovasc Electrophysiol 2006;17:1274 –8.
18. Squara F, Latcu DG, Massaad Y, Mahjoub M, Bun SS, Saoudi N. Contact force and
force-time integral in atrial radiofrequency ablation predict transmurality of
lesions. Europace 2014;16:660 –7.
19. Dixit S, Marchlinski FE, Lin D, Callans DJ, Bala R, Riley MP et al. Randomized ablation
strategies for the treatment of persistent atrial fibrillation: Rasta study. Circ Arrhythm
Electrophysiol 2012;5:287 –94.
20. Paisey J, Betts TR, De Bono J, Rajappan K, Tomlinson D, Bashir Y. Validation of coronary sinus activation pattern during left atrial appendage pacing for beat-to-beat assessment of mitral isthmus conduction/block. J Cardiovasc Electrophysiol 2010;21:
418 –22.
21. Maselli D, Guarracino F, Chiaramonti F, Mangia F, Borelli G, Minzioni G. Percutaneous mitral annuloplasty: an anatomic study of human coronary sinus and its relation
with mitral valve annulus and coronary arteries. Circulation 2006;114:377 –80.
22. Plass A, Valenta I, Gaemperli O, Kaufmann P, Alkadhi H, Zund G et al. Assessment of
coronary sinus anatomy between normal and insufficient mitral valves by multi-slice
computertomography for mitral annuloplasty device implantation. Eur J Cardiothorac
Surg 2008;33:583 – 9.
23. Piorkowski C, Hindricks G, Schreiber D, Tanner H, Weise W, Koch A et al. Electroanatomic reconstruction of the left atrium, pulmonary veins, and esophagus compared with the "true anatomy" on multislice computed tomography in patients
undergoing catheter ablation of atrial fibrillation. Heart Rhythm 2006;3:317 –27.
24. Latcu DG, Massaad Y, Mahjoub M, Squara F, Saoudi N. Peri-mitral flutter ablation
depends on mitral isthmus anatomy. Europace 2012;14(S1):136P–32.