JACC: Clinical Electrophysiology

May 2018 DOI: 10.1016/j.jacep.2018.03.007
PDF Article

Novel Electrocardiographic Criteria for Real-Time Assessment of Anterior Mitral Line Block
“V1 Jump” and “V1 Delay”

Heiko Lehrmann, Amir S Jadidi, Jan Minners, Juan Chen, Björn Müller-Edenborn, Reinhold Weber, Olaf Dössel, Thomas Arentz and Axel Loewe

Author + information

DOI: 
https://doi.org/10.1016/j.jacep.2018.03.007
Published By: 
JACC: Clinical Electrophysiology
Print ISSN: 
2405-5018
Online ISSN: 
2405-500X
History: 
  • Received November 13, 2017
  • Revision received March 1, 2018
  • Accepted March 6, 2018
  • Published online May 2, 2018.

Article Versions

Copyright & Usage: 
2018 The Authors

Author Information

  1. Heiko Lehrmann, MDa,∗ (heiko.lehrmann{at}universitaets-herzzentrum.de),
  2. Amir S Jadidi, MDa,
  3. Jan Minners, MD, PhDa,
  4. Juan Chen, MDa,
  5. Björn Müller-Edenborn, MDa,
  6. Reinhold Weber, MDa,
  7. Olaf Dössel, PhDb,
  8. Thomas Arentz, MDa and
  9. Axel Loewe, PhDb
  1. aDepartment of Cardiology and Angiology II, University Heart Center Freiburg/Bad Krozingen, Bad Krozingen, Germany
  2. bInstitute of Biomedical Engineering, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany
  1. ↵∗Address for correspondence:
    Dr. Heiko Lehrmann, Department of Cardiology and Angiology II, University Heart Center Freiburg/Bad Krozingen, Suedring 15, 79189 Bad Krozingen, Germany.

Abstract

Objectives This study hypothesized that P-wave morphology and timing under left atrial appendage (LAA) pacing change characteristically immediately upon anterior mitral line (AML) block.

Background Perimitral flutter commonly occurs following ablation of atrial fibrillation and can be cured by an AML. However, confirmation of bidirectional block can be challenging, especially in severely fibrotic atria.

Methods The study analyzed 129 consecutive patients (66 ± 8 years, 64% men) who developed perimitral flutter after atrial fibrillation ablation. We designed electrocardiography criteria in a retrospective cohort (n = 76) and analyzed them in a validation cohort (n = 53).

Results Bidirectional AML block was achieved in 110 (85%) patients. For ablation performed during LAA pacing without flutter (n = 52), we found a characteristic immediate V1 jump (increase in LAA stimulus to P-wave peak interval in lead V1) as a real-time marker of AML block (V1 jump ≥30 ms: sensitivity 95%, specificity 100%, positive predictive value 100%, negative predictive value 88%). As V1 jump is not applicable when block coincides with termination of flutter, absolute V1 delay was used as a criterion applicable in all cases (n = 129) with a delay of 203 ms indicating successful block (sensitivity 92%, specificity 84%, positive predictive value 90%, negative predictive value 87%). Furthermore, an initial negative P-wave portion in the inferior leads was observed, which was attenuated in case of additional cavotricuspid isthmus ablation. Computational P-wave simulations provide mechanistic confirmation of these findings for diverse ablation scenarios (pulmonary vein isolation ± AML ± roof line ± cavotricuspid isthmus ablation).

Conclusions V1 jump and V1 delay are novel real-time electrocardiography criteria allowing fast and straightforward assessment of AML block during ablation for perimitral flutter.

Key Words

Perimitral flutter is a common macro–re-entrant arrhythmia during and after ablation procedures in patients with persistent atrial fibrillation (AF) (1–3). In addition, some centers that follow a low-voltage/substrate-guided ablation approach in patients with persistent AF, apply “strategic” linear lesions independent of a macro–re-entrant tachycardia history (4). Three potential ablation strategies exist for a mitral line: an anterior mitral line (AML), a lateral mitral line or ethanol injection into the vein of Marshall ± additional radiofrequency ablation in the adjacent lateral wall (5,6). Compared with the AML, ablation of the lateral mitral isthmus is associated with an increased procedure time and the necessity for coronary sinus ablation with risk of coronary injury (5,7). The major disadvantage of the AML lies in the “transection” of the anterior left atrium (LA), with a severely delayed activation of the LA appendage (LAA) and reduced atrial transport function. Because of a high prevalence of preexistent low-voltage areas especially in the anterior and septal parts of the LA in comparison with the lateral LA (8), the AML is nevertheless usually the preferred linear ablation target in our hospital in these patients. However, assessment of bidirectional block with the classical criteria (9,10) can be complex and time consuming in these cases because of low amplitude signals in fibrotic tissue. We therefore searched for additional AML block criteria utilizing the surface electrocardiography (ECG) to overcome this limitation.

Methods

Patient selection and procedural details

From January 2014 to July 2017, data from consecutive patients scheduled for LA ablation procedures (paroxysmal and persistent AF), in whom the AML was targeted for ablation (because of perimitral flutter or after a low-voltage or substrate-guided ablation approach), were analyzed. Our ECG criteria were retrospectively developed (January 2014 to December 2016) and checked in a validation cohort (January to July 2017) (Figure 1). The study was approved by the Institutional Research Board. A detailed description of the AF ablation procedure was published previously (11). Ablation was performed using irrigated-tip ablation catheters (ThermoCool/ThermoCool Smarttouch, Biosense Webster, Irvine, California; CoolFlex/Tacticath, St. Jude Medical, St. Paul, Minnesota). Guided by preexistent low-voltage areas (<1.0 mV in sinus rhythm [SR]; <0.5 mV during AF) (12), the AML was applied between the anteroseptal mitral annulus and the left superior pulmonary vein, a roof line, or the right superior PV (Online Figure 1). The endpoint was electrical isolation of the PVs and bidirectional block of all applied linear lesions. Block was confirmed using previously published criteria (9,10): 1) widely spaced double potentials along the entire line under LAA pacing; 2) differential pacing on both sides of the line; and 3) mapping of the activation detour during pacing from either side of the line.

Figure 1
Figure 1

Study Population

SR = sinus rhythm.

V1 jump and V1 delay criteria

In case of SR at the time of AML completion, continuous pacing from the LAA was performed. The time interval between the pacing stimulus and the maximal positive deflection of the P-wave in ECG lead V1 was monitored (V1 delay). A sudden increase in this interval was termed “V1 jump” and correlation with synchronous bidirectional block of the AML was assessed. For patients in whom complete block was achieved, V1 jump was defined as the difference between V1 delay directly after block and the shortest V1 delay in the 5 preceding beats. If block could not be achieved, we defined V1 jump as the difference between V1 delay after and before the unsuccessful AML ablation in this group. Furthermore, concomitant changes in the P-wave morphology of the inferior ECG leads were analyzed.

In case of AML ablation during AF or atrial flutter, bidirectional block can precede termination to SR or occur simultaneously. Under these circumstances, the V1 jump criterion cannot be assessed. To overcome this limitation, we sought for an absolute V1 delay reflecting bidirectional block. The composition of the analyzed group is outlined in Figure 1.

Computer simulations

Because of the heterogeneity regarding ablation targets covered by the study population, we leveraged an in silico approach to mechanistically dissect the differential effects of the lesions. The simulation setup for excitation propagation and computation of body surface ECGs has been described and validated previously (13,14). In brief, an anatomical model of the torso including the heart, lungs, liver, and kidney was created based on magnetic resonance imaging data of a healthy subject (47 years of age). The biatrial model was augmented with rule-based fiber orientation, interatrial connections (via Bachmann’s bundle, the coronary sinus and posteriorly), and tissue labels to represent anisotropic and heterogeneous excitation propagation (14,15). Compared with Loewe et al. (14), monodomain tissue conductivity was homogeneously reduced to match V1 delay before AML ablation to that observed in the clinical study population. In silico ablation was modeled by rendering the tissue nonconductive. Starting from a reference setup without ablations, PV isolation was performed followed by AMLs that were connected with the left superior PV, the right superior PV, or an additional roof line. Besides successful lesions, scenarios with a gap in the inferior third of the mitral line were investigated. In a second set of simulations, additional cavotricuspid isthmus (CTI) ablation was considered (Online Figures 2 and 3).

Statistical analysis

Normal distribution of data was evaluated by visual assessment of histograms, probability plots (Q-Q plots) and by Kolmogorov-Smirnov or Shapiro-Wilk tests. Continuous variables following a normal distribution are given as mean ± SD and were analyzed using Student’s t test. In case of skewed distributions, continuous variables are presented as median with interquartile range (IQR) (first and third quartile). Statistical significance was assessed using the Mann-Whitney U test in case of 2 groups. Fisher’s exact test was used for assessing significance of categorical variables. The optimal cutoff value to detect AML block was determined by using receiver-operating characteristic (ROC) curve analysis. Parameter performance was assessed using the areas under the ROC curves.

Results

Perimitral flutter was treated by AML ablation in 129 patients (Figure 1). The presented ECG criteria were developed in a retrospective cohort that comprised 76 patients (January 2014 to December 2016) and were afterward evaluated in a validation cohort of 53 patients (January 2017 to July 2017). Table 1 summarizes the baseline characteristics of all patients and the 2 cohorts. The AML was connected from the anteroseptal mitral annulus to the left superior PV, a roof line, or the right superior PV in 14%, 44%, and 42% of the cases, respectively. Bidirectional AML block was achieved in 85% of the procedures. In patients in which AML block could not be achieved (n = 19), crossover to a lateral mitral line was performed in only 4 patients (necessary coronary sinus ablation in 2 patients). In 29% of cases, ablation was performed during LAA pacing without ongoing AF or atrial flutter leading to direct visibility of the immediate block on the surface ECG. Additional linear lesions were performed in 81% of cases (CTI ablation in 46%, roof line in 62%). Follow up-over 13 ± 10 months was performed by the referring cardiologists. Twelve of 129 (9%) patients presented with recurrence of sustained flutter and had a redo ablation procedure. In 8 of 12 patients, a gap in the AML was identified including 2 patients in whom AML block was not achieved during the index procedure. Systematic follow-up including long-term Holter monitoring or tele-ECG was not performed, as the focus of this study is the assessment of acute AML block. Thus, an overall long-term success rate of the ablation strategy including the AML cannot be given here.

View this table:
Table 1

Baseline Characteristics

V1 jump criterion

In 52 cases, AML ablation was conducted in SR during LAA pacing. Figure 2 exemplarily illustrates the characteristic P-wave changes upon AML block. The immediate mitral line block, as evidenced by the intracardiac tracings, is accompanied by an abrupt increase in the time interval between the stimulus and the peak in V1 (V1 jump). Additionally, an inversion of the inferior P-wave polarity from a purely positive to negative-positive is apparent. Figure 3A depicts the typical V1 jump and inferior P-wave morphology changes during AML block in a different patient. The activation map after AML block (Figure 3B) explains the observed ECG changes. The delayed peak in V1 under LAA pacing upon AML block is explained by the pronouncedly delayed activation of the right atrial free wall (Figure 3: pink color next to right atrial appendage). The negative P waves in the inferior leads result from the inferior-to-superior–oriented activation front in the left posterior wall, the septum, and the right atrium. Figure 4 depicts the described P-wave morphology changes (before AML block vs. after block) (Figure 4A vs. Figure 4B) in 5 additional patients. Furthermore, it shows the results of a biatrial computer simulation (P-wave and activation map). A blocked AML, between the anteroseptal mitral annulus and the left superior PV after PV isolation in this scenario, reproduced the clinically observed P-wave changes and the clinical activation map mechanistically. Similar results were observed with different AML localizations (roof line or right superior PV) and are presented in the supplemental material (Online Figures 2 and 3).

Figure 2
Figure 2

V1 Jump and Intracardiac Tracing During Acute AML Block

Surface electrocardiography and intracardiac tracings of a patient with observable anterior mitral line (AML) block. Between the third and fourth beats, AML block occurs, which is immediately apparent in an increasing intracardiac delay (116 to 182 ms, blue arrows). Concurrently, a V1 jump occurs (pacing stimulus to P-wave peak in V1 from 154 to 230 ms; red arrow) with a P-wave polarity change in the inferior leads (green arrows).

Figure 3
Figure 3

V1 Jump and Clinical Activation Map

(A) Surface electrocardiography (in a different patient than Figure 2) with anterior mitral line block. Block is directly discernable between the third and fourth beats based on V1 jump (red arrow) and a P-wave polarity change in the inferior leads (green arrows). (B) Activation map of the left and right atria after anterior mitral line block. Note the latest activation in the right atrial free wall adjacent to the appendage explaining the late positive P-wave peak in V1. Terminal forces in both atria are directed from inferior to superior (white arrows) explaining the terminal P-wave negativity in the inferior leads.

Figure 4
Figure 4

Surface Electrocardiography (V1 Delay) Before and After Anterior Mitral Line Block

The 12-lead surface electrocardiography during left atrial appendage pacing in 5 clinical cases and the computer-simulated P-wave (“Sim”) (A) before and (B) after anterior mitral line (AML) block. Note the inferior P-wave negativity (green arrows) and the V1 delay prolongation after block (red arrows) with a profound activation change and a resultant very late depolarization of the right atrial free wall. P-wave simulation resembles the clinical P waves in both scenarios (pacing stimulus and QRS complexes are not part of the simulated signals). The simulated activation maps in the bottom panels resolve the details of the biatrial activation patterns before and after block.

Figure 5A shows a scatterplot of the V1 jump criterion in the subgroup of patients with directly visible AML block (n = 38) and the patients in whom block was not achieved during the procedure (n = 14).

Figure 5
Figure 5

V1 Jump Scatterplot and ROC Curve

(A) V1 jump scatterplot of patients in the retrospective cohort, the validation cohort, and all patients (3 columns). Each column depicts patients without anterior mitral line block on the left (circles) and patients with AML block on the right (diamonds). The green dashed line indicates the proposed cutoff value of 24 ms based on the receiver-operating characteristic (ROC) curve of the retrospective cohort. The red solid line shows the suggested cutoff value of 30 ms based on (B) the ROC curve of all patients. In A, whisker plots show mean and standard deviation in the left 2 columns (normal distribution) and median and interquartile range in the right column (skewed distribution). AUC = area under the receiver-operating characteristic curve.

Initial criteria were developed on the basis of a retrospective cohort (Figure 5A, left column) (n = 26). Mean V1 jump (increase in V1 delay) was 66 ± 36 ms and 13 ± 10 ms with and without AML block, respectively (p < 0.0001). ROC curve analysis (data not shown) identified a V1 jump of ≥24 ms (dashed green line in Figure 5A) as an appropriate predictor of AML block with sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of 94%, 80%, 88%, and 89%, respectively. The criteria were subsequently examined in a validation cohort (Figure 5A, middle column) (n = 26). In this group, mean V1 jump was 59 ± 17 ms and 5 ± 3 ms with and without AML block, respectively (p < 0.0001). Taking all patients together (Figure 5A, right column) (n = 52) a median V1 jump of 57 ms (IQR: 40 to 78 ms) and 8 ms (IQR: 2 to 16 ms) with and without AML block was observed, respectively (p < 0.0001). ROC curve analysis in all patients (Figure 5B) suggested a V1 jump of ≥30 ms as the most appropriate predictor of AML block with sensitivity, specificity, PPV, and NPV of 95%, 100%, 100%, and 88%, respectively.

V1 delay criterion

AML block is not always immediately assessable (e.g., because the line is already bidirectionally blocked after conversion of perimitral flutter to SR). Therefore, we studied the significance of the absolute V1 delay during LAA pacing (n = 110 patients with AML block; n = 70 patients before or without AML block) (Figure 1). Figure 6A depicts V1 delay scatterplots in the retrospective, and validation cohort and in all patients divided into blocked and nonblocked AML. In the retrospective cohort, the median V1 delay was 156 ms (IQR: 150 to 185 ms) in patients without or before AML block (n = 41) (Figure 6A, left column). On the other hand, patients with bidirectional AML block (n = 64) showed a median V1 delay of 243 ms (IQR: 219 to 274 ms; p < 0.0001). ROC curve analysis identified a V1 delay of 194 ms as the cutoff with the highest sensitivity (97%) and specificity (85%) for assessment of AML block in these patients (data not shown). PPV and NPV for this cutoff were 91% and 95%, respectively.

Figure 6
Figure 6

V1 Delay Scatterplot and ROC Curve

(A) V1 delay scatterplot of patients in the retrospective cohort, the validation cohort, and all patients. Each column depicts patients without anterior mitral line block on the left (circles) and patients with anterior mitral line block on the right (diamonds). The green dashed line indicates the proposed cutoff value of 194 ms based on the ROC curve of the retrospective cohort. The red solid line shows the suggested cutoff value of 203 ms based on (B) the ROC curve of all patients. In A, whisker plots show mean and standard deviation in the middle column (normal distribution) and median and interquartile range in the right and left column (skewed distribution), respectively. Abbreviations as in Figure 5.

The V1 delay criterion was also tested in the validation cohort (Figure 6A, middle column). Mean V1 delay in patients without or before AML block was 179 ± 32 ms (n = 29). In case of AML block, mean V1 delay was significantly increased to 242 ± 27 ms (n = 46; p < 0.0001).

Analyzing all patients (Figure 6A, right column), a median V1 delay of 243 ms (IQR: 220 to 264 ms) with AML block was observed (n = 110). The corresponding median intracardiac mitral line delay was 189 ms (IQR: 174 to 214 ms; measured interval from LAA stimulus to the electrogram recorded at the septal aspect of the AML). On the other hand, patients without or before AML block (n = 70) showed a median V1 delay of 164 ms (IQR: 152 to 193 ms; p < 0.0001) and a median intracardiac delay of 127 ms (IQR: 107 to 138 ms). ROC curve analysis in all patients (Figure 6B) suggested a V1 delay of ≥203 ms as the most appropriate predictor of AML block with sensitivity, specificity, PPV, and NPV of 92%, 84%, 90%, and 87%, respectively.

Characteristic P-wave morphology with and without CTI block

AML block during LAA pacing does not only lead to P-wave changes in lead V1 (V1 jump or V1 delay) but also to a characteristic polarity change of the inferior P-wave morphology. This change is highly dependent on concomitant CTI block (n = 7). Figure 2 shows a typical tracing of a patient with a previous CTI block during LAA pacing. Subsequent AML block lead to a slightly negative vector in the inferior P waves accompanied by a major terminal positivity in the same leads. Figure 3, on the other hand, depicts a typical tracing in a patient without CTI block. AML block also resulted in a negative inferior P-wave vector but the major terminal positivity in the inferior leads was lacking. Figure 7 summarizes these results exemplarily in 5 patients in whom CTI ablation was performed after AML block and in a computer simulation. Before CTI block, inferior P-wave morphology showed a negative vector as well as a slight terminal positivity in some patients and the computer simulation (Figure 7A). This terminal positivity significantly increased or arose after CTI block in all patients and the simulation. Moreover, the terminal positivity after CTI block was accompanied by a reduction or loss of the characteristic negative inferior P-wave polarity. The simulated activation maps elucidate the mechanism behind these changes: without CTI block, the terminal P-wave vector of the inferior leads is balanced between the inferosuperior “trans-CTI” activation and the superoinferior activation causing latest activation in the RA free wall (Figure 7A). Depending on the conduction velocity distribution and the specific anatomy, this activation pattern leads to a more or less negative inferior P-wave vector. In case of CTI block, the inferosuperior “trans-CTI” activation is blocked, leading to an unbalanced superior-inferiorly directed activation of the lateral free wall and a resultant terminal positive P-wave vector in the inferior leads (Figure 7B).

Figure 7
Figure 7

Surface Electrocardiography (V1 Delay) After Anterior Mitral Line Block With and Without CTI Block

The 12-lead surface electrocardiography during left atrial appendage pacing in 5 clinical cases and the computer-simulated P-wave (“Sim”) after anterior mitral line block (A) with and (B) without cavotricuspid isthmus (CTI) block. Note the increasing terminal inferior P-wave positivity after CTI block (red arrows before and after block). The simulated high-resolution activation maps explain the observed P-wave changes: after CTI block, the latest atrial activation shifts from the right atrial free wall to the lateral CTI (next to the line) with a resulting unopposed inferiorly directed P-wave vector. The P-wave simulation resembles the clinical P waves in both scenarios (pacing stimulus and QRS complexes are not part of the simulated signals).

Computer simulations

To systematically study the effect of the different AMLs (mitral annulus to left superior PV or roof line or right superior PV, with or without additional roof line or CTI ablation) on V1 jump and V1 delay under well-controlled conditions and to complement the limited amount of clinical cases in each category, computer simulations were utilized. In concordance with the clinical data, V1 delay was unaffected by any combination of PV isolation and a roof line (simulated V1 delay: 149 to 150 ms). In case of a conducting gap in the AML, simulated V1 delay was 169 ms, 185 ms, and 189 ms for mitral lines to left superior PV, roof line, and right superior PV, respectively. With complete AML block, these values increased by a mean of 88 ms to 260 ms, 273 ms, and 274 ms, respectively. Additional CTI ablation caused negligible extra V1 delay of <1 ms. All simulated scenarios were correctly classified using both V1 jump and V1 delay criteria with ample safety margins to the proposed cutoff values (V1 jump: 55 ms margin; V1 delay: 57 ms margin).

Discussion

Main findings

We established 2 novel and easily applicable real-time surface ECG criteria to assess AML block during LA ablation procedures: V1 jump and V1 delay. Both criteria are based on changes in ECG lead V1 during LAA pacing. These new criteria provide rapid, robust, and accurate tools for the assessment of AML block with high sensitivity, specificity, and PPV and NPV.

The study was conducted in an AF ablation cohort with a total of 129 patients. The criteria were developed in a retrospective cohort (n = 76) and tested in a validation cohort (n = 53).

Nevertheless, it is important to note that AML patients represent a heterogeneous group due to the variability in anatomical ablation targets. Connections between the anterior or anteroseptal mitral annulus and the right superior PV, the left superior PV, or a concomitant roof line are conceivable and relevant in clinical practice. Often, additional linear lesions are performed in these patients (e.g., a CTI line). To evaluate a novel AML block marker such as the V1 jump or V1 delay criteria, a huge number of patients need to be studied in each group to draw definite conclusions. Therefore, we combined the clinical data with biatrial computer simulations to investigate each scenario mechanistically and to resolve the underlying mechanisms. In summary, we found that both criteria are independent of AML localization and additional ablation lines (e.g., concomitant roof or CTI lines).

V1 delay

Considering both criteria separately, the main strength of the V1 delay criterion lies in its ability to rapidly evaluate AML block during an ablation procedure after termination to SR. Thus, it is much faster than time-consuming multipoint activation mapping, differential pacing, and a search for double potentials, which can be cumbersome due to very low-voltage signals as a result of extensive atrial fibrosis (9,10). In contrast, our new approach, which consists of simple LAA pacing and surface ECG monitoring, is independent of intracardiac signal amplitude and can therefore be rapidly applied even under these circumstances. A limitation of the V1 delay criterion is the remaining overlap between the groups with and without AML block. However, the proposed cutoff value of 203 ms serves as a good discriminator in the majority of patients according to our data.

V1 jump

The V1 jump criterion provides an easily applicable real-time tool to assess instantaneous AML block during ablation in SR with excellent test characteristics. The V1 jump definition in the “no block” cohort was set as the difference in V1 delay directly before and after the attempted AML because no real “jump,” but rather a continuously increasing delay was observed in these patients. Due to this strict definition, sensitivity and NPV are presumably even higher in practice.

Comparison with previous publications

In 2015, Huemer et al. (16) reported their experience with AML ablation in SR. They observed an LAA conduction jump of at least 50 ms at the time of successful AML block (mean 81 ms). Only 1 patient with a previous AML ablation and subsequent conduction recovery showed an LAA jump of only 30 ms. These data are in accordance with our results with a median V1 jump of 57 ms. Furthermore, an additional roof line did not alter the LAA conduction jump as observed in our study. Nevertheless, a couple of limitations apply to the previous study (16): (A) The reported LAA conduction jump is only applicable if AML block occurs during ablation in SR. In contrast, we provide a V1 delay cutoff serving as an indicator of block even in cases when the AML is already blocked after termination to SR. (B) The previous work is limited to an AML connecting to the left superior PV whereas our clinical data and computer simulations cover all AML locations and combinations with other relevant lines (roof line, CTI ablation).

Study limitations

Several limitations need to be acknowledged. First, the reported V1 jump and V1 delay criteria can only assess unidirectional AML block (from the lateral to the septal side), although all patients in the presented cohort had bidirectional block confirmed by the “classical criteria.” Second, the classical criteria were used as the gold standard for AML block assessment. Although their use is fairly straightforward, despite all care, a misclassification can never be completely ruled out. Third, 2 potential limitations of the V1 jump and V1 delay criteria were identified through simulations: 1) A completely isolated block of Bachmann’s bundle without AML block led to an activation map and V1 delay (229 ms) similar to the AML block scenario; and 2) An AML connecting the mitral annulus to the right superior PV but not affecting Bachmann’s bundle did not increase V1 delay upon AML block (149 ms) (Online Figures 2 and 3). Although generally possible, it does not seem to be a desired procedural endpoint because of the risk for biatrial tachycardia in these cases (17). Of note, both these scenarios were not observed in the clinical cases. Fourth, the diagnostic thresholds together with the related values for sensitivity and specificity as well as NPV and PPV need to be considered as specific for our study population. In this population, the desired outcome (AML block) was achieved in 85% of patients. The previously mentioned measures might differ in a population with a lower AML block prevalence.

Clinical implications

We currently adhere to the following protocol in our laboratory: If an AML is attempted during ongoing perimitral flutter, V1 delay is measured immediately after conversion to SR to rapidly check for AML block. Hereby, a V1 delay of <203 ms makes AML block unlikely and we immediately look for gap signals on the whole line without performing differential pacing or activation mapping. In case of a remaining AML gap, the V1 jump criterion (≥30 ms) can then be used as a real-time indicator of AML block. Moreover, V1 jump can be applied in patients with a history of perimitral flutter or after a substrate-guided ablation approach with application of a “strategic” AML (4), who are ablated during SR. Its main advantage lies in its real-time capability of AML block assessment and is therefore comparable to the coronary sinus activation reversal during LAA pacing with lateral mitral isthmus block (10). Real-time AML block detection is particularly valuable during ablation, which often obscures the intracardiac signals due to radiofrequency noise. In these cases, instantaneous AML block can still be easily assessed by the V1 jump criterion. This allows for additional ablation at this crucial point to consolidate AML block. Furthermore, reversion of V1 jump (decreasing V1 delay), which can easily be recognized on the surface ECG, serves as a real-time indicator of conduction recovery across the AML during the procedure (Online Figure 4). In addition, a simultaneous negativity in the inferior leads can serve as an additional block marker and is easily visualized during the procedure. As mentioned previously, this polarity reversal may be attenuated or even reversed by CTI ablation.

Finally, 3 points should be emphasized. First, the new criteria are not meant to be exclusive markers for block, but rather should be used in conjunction with the classical block criteria (9,10). Second, the current study is not intended to promote the AML approach above the lateral approach. Both ablation strategies have their advantages and disadvantages. The major disadvantage of the AML is the “transection” of the anterior LA with a substantially delayed electrical LAA-activation (in our cohort: median interval from LAA potential to QRS onset −10 ms (IQR: −44 to +8 ms; data not shown), which potentially results in an impaired atrial transport function. Moreover, it must be emphasized that additional ablation of the lateral wall (e.g., crossover to lateral mitral line in case of AML failure) should be avoided to prevent accidental electrical LAA isolation, which is associated with higher thromboembolic risk despite therapeutic oral anticoagulation (18). Third, the current study does not suggest a substrate-guided linear ablation approach in patients without perimitral flutter. Whether this improves long-term outcome has to be shown in prospective multicenter trials. On the contrary, we think that purely empiric linear lesion sets should be avoided in patients treated for AF, as was shown (e.g., in the STAR AF II [Substrate and Trigger Ablation for Reduction of Atrial Fibrillation Trial-Part II] trial) (19). An important aspect is conduction recovery of linear lesions during follow-up with a resultant “proarrhythmic” effect (20).

Conclusions

With the novel V1 jump and V1 delay criteria, AML block can easily be assessed during LAA pacing by monitoring the surface ECG. V1 delay primarily serves as a sensitive (92%), rapid rule-out criterion for AML block after termination of perimitral flutter to SR. On the other hand, V1 jump can be used as a real-time indicator of AML block in patients with a remaining AML gap with a high specificity and PPV of 100%. In combination, these criteria provide a rapid and accurate tool to evaluate AML block intraprocedurally.

Perspectives

COMPETENCY IN MEDICAL KNOWLEDGE: Our novel surface ECG criteria (V1 jump and V1 delay) provide the practicing electrophysiologist with a powerful and easy-to-use tool for AML block assessment during LA ablation procedures. These new criteria are rapidly applicable, whereas proof of AML block with the classical block criteria can be challenging and time consuming, especially in severely diseased or fibrotic left atria due to low-voltage signals. The new criteria are not meant to be exclusive, but rather increase the electrophysiologist’s armamentarium for block assessment (clinical competency: “procedural skills”). Furthermore, V1 jump, a real-time detector of AML block, is useful even if intracardiac signals are obscured due to radiofrequency application noise or in case of severe atrial fibrosis. This allows for additional ablation time on a crucial atrial tissue area for long-term consolidation of AML block (clinical competency: “improved patient care”).

TRANSLATIONAL OUTLOOK: Our collaborative work shows that combining clinical research with computational modeling allows easier understanding of an observed clinical phenomenon. Furthermore, hypothesis testing can be facilitated even in a complex patient cohort (e.g., patients with different anatomical AML lines or additional ablation targets). Nevertheless, further research is desirable to confirm our results in an even bigger patient cohort.

Acknowledgment

The authors would like to thank Dr. Gunnar Seemann for initiating the contact leading to the collaborative research presented here.

Appendix

Online Figures 1–4 [S2405500X1830210X_mmc1.pdf]

Footnotes

  • This study received financial support from the “Deutsche Forschungsgemeinschaft (DFG)” through CRC 1173. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.

  • All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.

Abbreviations and Acronyms
AF
atrial fibrillation
AML
anterior mitral line
CTI
cavotricuspid isthmus
ECG
electrocardiography
IQR
interquartile range
LA
left atrium/atrial
LAA
left atrial appendage
NPV
negative predictive value
PPV
positive predictive value
PV
pulmonary vein
ROC
receiver-operator characteristic
SR
sinus rhythm
  • Received November 13, 2017.
  • Revision received March 1, 2018.
  • Accepted March 6, 2018.

References

  1. ↵
    1. Page R.L.,
    2. Joglar J.A.,
    3. Caldwell M.A.,
    4. et al.
    (2016) 2015 ACC/AHA/HRS guideline for the management of adult patients with supraventricular tachycardia: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 67:e27–e115.
    OpenUrlFREE Full Text
  2. ↵
    1. Matsuo S.,
    2. Wright M.,
    3. Knecht S.,
    4. et al.
    (2010) Peri-mitral atrial flutter in patients with atrial fibrillation ablation. Heart Rhythm 7:2–8.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Ammar S.,
    2. Luik A.,
    3. Hessling G.,
    4. et al.
    (2015) Ablation of perimitral flutter: acute and long-term success of the modified anterior line. Europace 17:447–452.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Kircher S.,
    2. Arya A.,
    3. Altmann D.,
    4. et al.
    Individually tailored vs. standardized substrate modification during radiofrequency catheter ablation for atrial fibrillation: a randomized study. Europace, [E-pub ahead of print]. 2017 Nov 21.
  5. ↵
    1. Huemer M.,
    2. Wutzler A.,
    3. Parwani A.S.,
    4. et al.
    (2015) Comparison of the anterior and posterior mitral isthmus ablation lines in patients with perimitral annulus flutter or persistent atrial fibrillation. J Interv Card Electrophysiol 44:119–129.
    OpenUrl
  6. ↵
    1. Baez-Escudero J.L.,
    2. Morales P.F.,
    3. Dave A.S.,
    4. et al.
    (2012) Ethanol infusion in the vein of Marshall facilitates mitral isthmus ablation. Heart Rhythm 9:1207–1215.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Wittkampf F.H.,
    2. van Oosterhout M.F.,
    3. Loh P.,
    4. et al.
    (2005) Where to draw the mitral isthmus line in catheter ablation of atrial fibrillation: histological analysis. Eur Heart J 26:689–695.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Yagishita A.,
    2. De Oliveira S.,
    3. Cakulev I.,
    4. et al.
    (2016) Correlation of left atrial voltage distribution between sinus rhythm and atrial fibrillation: identifying structural remodeling by 3-D electroanatomic mapping irrespective of the rhythm. J Cardiovasc Electrophysiol 27:905–912.
    OpenUrl
  9. ↵
    1. Shah D.,
    2. Haissaguerre M.,
    3. Takahashi A.,
    4. Jais P.,
    5. Hocini M.,
    6. Clementy J.
    (2000) Differential pacing for distinguishing block from persistent conduction through an ablation line. Circulation 102:1517–1522.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Jais P.,
    2. Hocini M.,
    3. Hsu L.F.,
    4. et al.
    (2004) Technique and results of linear ablation at the mitral isthmus. Circulation 110:2996–3002.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Arentz T.,
    2. Weber R.,
    3. Burkle G.,
    4. et al.
    (2007) Small or large isolation areas around the pulmonary veins for the treatment of atrial fibrillation? Results from a prospective randomized study. Circulation 115:3057–3063.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Jadidi A.S.,
    2. Lehrmann H.,
    3. Keyl C.,
    4. et al.
    (2016) Ablation of persistent atrial fibrillation targeting low-voltage areas with selective activation characteristics. Circ Arrhythm Electrophysiol 9:e002962.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Krueger M.W.,
    2. Seemann G.,
    3. Rhode K.,
    4. et al.
    (2013) Personalization of atrial anatomy and electrophysiology as a basis for clinical modeling of radio-frequency ablation of atrial fibrillation. IEEE Trans Med Imaging 32:73–84.
    OpenUrl
  14. ↵
    1. Loewe A.,
    2. Krueger M.W.,
    3. Holmqvist F.,
    4. Dossel O.,
    5. Seemann G.,
    6. Platonov P.G.
    (2016) Influence of the earliest right atrial activation site and its proximity to interatrial connections on P-wave morphology. Europace 18:iv35–iv43.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Wachter A.,
    2. Loewe A.,
    3. Krueger M.,
    4. Dössel O.,
    5. Seemann G.
    (2015) Mesh structure-independent modeling of patient-specific atrial fiber orientation. Curr Direct Biomed Eng 1:409–412.
    OpenUrl
  16. ↵
    1. Huemer M.,
    2. Wutzler A.,
    3. Parwani A.S.,
    4. et al.
    (2015) Left atrial appendage conduction jump for real-time evaluation of conduction block over the anterior mitral annulus line. J Cardiovasc Electrophysiol 26:730–736.
    OpenUrl
  17. ↵
    1. Mikhaylov E.N.,
    2. Mitrofanova L.B.,
    3. Vander M.A.,
    4. et al.
    (2015) Biatrial tachycardia following linear anterior wall ablation for the perimitral reentry: incidence and electrophysiological evaluations. J Cardiovasc Electrophysiol 26:28–35.
    OpenUrl
  18. ↵
    1. Rillig A.,
    2. Tilz R.R.,
    3. Lin T.,
    4. et al.
    (2016) Unexpectedly High incidence of stroke and left atrial appendage thrombus formation after electrical isolation of the left atrial appendage for the treatment of atrial tachyarrhythmias. Circ Arrhythm Electrophysiol 9:e003461.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Verma A.,
    2. Sanders P.,
    3. Macle L.,
    4. et al.
    (2012) Substrate and Trigger Ablation for Reduction of Atrial Fibrillation Trial-Part II (STAR AF II): design and rationale. Am Heart J 164:1–6.e6.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Sawhney N.,
    2. Anand K.,
    3. Robertson C.E.,
    4. Wurdeman T.,
    5. Anousheh R.,
    6. Feld G.K.
    (2011) Recovery of mitral isthmus conduction leads to the development of macro-reentrant tachycardia after left atrial linear ablation for atrial fibrillation. Circ Arrhythm Electrophysiol 4:832–837.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

Toolbox

Print PDF
Email
Citation

Metrics

Advertisement

Article Outline

Figures in Article

Figures

  • Figure1
  • Figure 1
  • Figure 2
  • Figure 3
  • Figure 4
  • Figure 5
  • Figure 6
  • Figure 7

Tables

Supplementary Materials

Similar Articles

Advertisement