Role of surface electrocardiogram precordial leads in localizing different anatomic sites of ectopic atrial tachycardia arising from lower right atrium in pediatric population

Abstract

Background

The study was designed to examine P wave morphology (PWM) in precordial leads (V₁–V₆) during ectopic atrial tachycardia (EAT) originating from low right atrium (RA) to identify the anatomic sites of these foci in children.

Methods

Twenty‐three consecutive pediatric patients (56% females, mean age 8.5 ± 2.5) with EAT originating from the low RA underwent detailed atrial endocardial activation mapping and radiofrequency ablation. PWM during EAT was analyzed using standard 12‐lead ECG in relation to successful ablation sites in RA.

Results

Ectopic atrial tachycardia originated from coronary sinus ostium (CSo) in 12 patients, nonseptal tricuspid annulus (TA) in five, lower crista terminalis (CT) in three and lower free wall in three. In lead V₁, PWM showed a positive pattern during EAT originating from CSo (8/12) [91.7% sensitivity, 100% specificity, 100% positive predictive value (PPV), 100% negative predictive value (NPV)]. A negative pattern was observed in EAT originating from lower free wall (1/3) and nonseptal TA (5/5) [50% sensitivity, 100% specificity, 100% PPV, 75% NPV], while isoelectric pattern was in EAT originating from lower CT (3/3) [100% sensitivity, 100% specificity, 100% PPV, 100% NPV]. In leads V₃–V₆, PWM showed a negative pattern in at least two consecutive leads during EAT from CSo (12/12), nonseptal TA (5/5) and lower free wall (3/3) while it was positive in EAT originating from lower CT (3/3) [100% sensitivity, 95% specificity, 75% PPV and 100% NPV].

Conclusions

P wave morphology in precordial leads can help differentiate the anatomic sites of EAT from lower RA with high PPVs and NPVs.

1. INTRODUCTION

Ectopic atrial tachycardia (EAT) is the most common chronic supraventricular tachycardia (SVT) in children accounting for up to 10% of SVT in patients < 30 years (Porter et al., 2004) and is often resistant to medical therapy, with radiofrequency ablation (RFA) being a preferred treatment option for children with refractory EAT (Case & Gillette, 1993). Clustering of foci of atrial tachycardia at specific anatomic structures has been well‐recognized (Badhwar et al., 2005; Kistler et al., 2006; Roberts‐Thomson et al., 2007; Yamada et al., 2007). Majority of EAT originate from right atrium (RA), most frequently along crista terminals (CT) in two‐thirds of cases (Kalman et al., 1998). Analysis of P wave morphology (PWM) in 12‐lead electrocardiogram (ECG) during EAT was used as a noninvasive tool for identifying the anatomical sites of these foci (Kistler et al., 2006) and many algorithms were published based on these analyses for preliminary localization of EAT (Kistler et al., 2006). Although, these algorithms were successful in identifying different RA and LA foci but it fell short in discriminating foci arising from lower RA. However, PWM analysis in precordial leads especially V₃–V₆ has not been sufficiently addressed in previous studies (Kistler et al., 2006) except for only one study (Huo et al., 2013) which showed the value of these leads to provide additional information that help distinguish between different Low RA foci in adult population.

Based on these data, this study was designed to further study and establish the value of PWM in precordial leads V₁–V₆ during EAT originating from lower RA to accurately identify the anatomic sites of its foci in pediatric population.

2. METHODS

2.1. Study population

This was a retrospective study which initially recruited 31 consecutive pediatric patients who underwent radiofrequency catheter ablation (RFCA) of EAT with RA origins at our EP lab in Ain shams university hospitals during the period from 2009 till 2014. Kistler et al. (2006), ECG algorithm was used for initial baseline localization of focus site based on the analysis of PWM in leads V₁ and aVL during tachycardia and patients who had EAT with RA origins based on the algorithm, were initially recruited this study.

Then EAT with lower RA origin as confirmed using either three‐dimensional (3D) electroanatomical (EA) activation mapping or conventional fluoroscopy guided activation mapping during the RFCA procedure were found in 23 out of the 31 patients (74%) who were included in our analysis while eight patients were excluded from the study as their EAT was originating from other atrial sites (six superior CT EAT, one EAT from right atrial appendage and one EAT from right inferior pulmonary vein). All Patients had 12‐lead surface ECG documentation of EAT that was resistant to medical treatment with at least one antiarrhythmic agent.

Patients were excluded if they were found to have other forms of atrial or ventricular arrhythmias or their EAT was confirmed to be from left atrium either by ECG or during electrophysiological study.

2.2. Mapping

The study was performed under conscious sedated state or general anesthesia after obtaining informed written consent from their parents and all anti‐arrhythmic drugs were discontinued at least five half lives before the study.

Mapping for localization and ablation of the tachycardia was performed using either Conventional fluoroscopy guided mapping or 3D EA mapping using Carto^R XP system (Biosense Webster, Diamond Bar, CA, USA).

Simultaneous surface ECG and bipolar intracardiac electrogram (EGM) recordings (filtered between 30 and 500 Hz) were amplified and displayed using the Prucka CardioLab System (GE Medical Systems, Milwaukee, WI, USA).

2.2.1. Conventional fluoroscopy guided mapping

Two decapolar catheters were introduced, one via the left subclavian vein and placed in the coronary sinus (CS) and the other via the femoral vein and placed around tricuspid annulus for mapping of RA. One quadripolar catheter was introduced via femoral vein puncture and placed in His‐bundle region. Local activation times obtained from catheters positioned at routine sites were assessed. Then, detailed activation mapping was achieved by the ablation catheter positioned within the region of presumed early activation to identify the earliest activation time relative to the onset of the P wave on surface ECG. To measure activation time to initial onset of the P wave, the 12‐lead ECG was analyzed carefully during a period of AV block.

3. 3D ELECTROANATOMICAL MAPPING

3D EA mapping was performed using Carto XP system (Biosense Webster). An 8F Navistar catheter with a 3.5 mm distal tip electrode was used for mapping and ablation (Navistar Thermocool, Biosense Webster). Proximal CS bipolar EGM was used as time reference. Both activation and isochronal maps were created during tachycardia.

When EAT was absent or intermittent during the procedure, we used provocative pharmacological maneuvers in the form of Isoproterenol infusion in a dose of 1–4 mg/min and then programmed electrical stimulation (PES) was performed.

A focal mechanism was suggested by a radial spread of activation from the discrete region (point) of atrial myocardium and by a total atrial activation time that is relatively short, so that large portions of the cycle are devoid of electrical activity, even when recording from both atria (Cosio, Goicolea, & Pastor, 2003). Detailed (high density) mapping was then carried out around the site of earliest activation to ensure no earlier activation can be found around this region (Liuba & Walfridsson, 2008) either by conventional or 3D EA mapping.

Activation time was measured from the onset of the first sharp component of the bipolar electrogram on the distal mapping catheter to the earliest deflection of the P wave on the surface ECG (Huo et al., 2013).

3.1. Radiofrequency ablation and follow‐up

The target site of ablation was determined using the combination of the earliest bipolar activation and the shape of the unipolar electrogram (QS pattern).

Radiofrequency catheter ablation (RFCA) was performed in the RA with continuous temperature feedback control of the power output to achieve a target temperature of 48°C. The maximum power used was 30W for a maximum of 60 s. An acute procedural success was defined by the inability to induce the tachycardia 20 min after ablation despite the aggressive pacing and PES protocols and the use of IV isoproterenol. All patients were followed in their referring clinics in order to assess the recurrence of symptoms or documented tachycardia.

3.2. Determination of site of RA EAT

1. An AT was considered to arise from the lower CT when earliest activation was mapped in this region with the aid of fractionated electrograms and the anatomical position was tagged on the 3D EA map and right anterior oblique (RAO) fluoroscopic view and ablation in this region successfully eliminated the tachycardia.

2. An AT was considered to arise from the CS ostium (CSo) when earliest activation was recorded around the CSo and ablation within 1 cm of this region successfully eliminated the tachycardia.

3. An AT was considered to arise from the Tricuspid annulus (TA) when ablation catheter was positioned in an annular location when viewed in right and left anterior oblique fluoroscopic views with characteristic annular motion of the catheter tip and exclusion of sites around the CSo and parahisian region. The different sectors of the TA were described using anatomic terminology contained in published guidelines (Cosio et al., 1999).

4. An AT was considered to arise from the lower free wall when earliest activation was mapped in the region between nonseptal TA and lower CT.

3.3. P wave analysis

The PWM on the surface ECG was assessed carefully during EAT for preliminary localization using Kistler et al. (2006) algorithm. A more accurate assessment of PWM was made during periods of AV block during EAT. A precise assessment of PWM in precordial leads most specifically V₃–V₆ was made in all patients and the acquired data were analyzed based on the electrophysiologically determined location of EAT foci.

P waves were described on the basis of the deviation from baseline during the TP interval as being: (i) completely positive (+), (ii) completely negative (−), (iii) biphasic: when there were both positive and negative (+/− or −/+) deflections from baseline, (iv) isoelectric: when there were no P wave deflections from baseline > 0.05 mV (Huo et al., 2013).

3.4. Statistical analysis

All data were revised and statistically analyzed using SPSS statistical package version 15 (IBM, Chicago, IL, USA). Continuous data are expressed as Mean and standard deviation. Population proportions were presented as a percentage. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated according to standard definitions.

4. RESULTS

4.1. Patient characteristics

There were 10 males (44%) and 13 females (56%) ranging in age from 3 to 12 years with a mean age of 8.6 ± 2.6 years. The mean tachycardia cycle length (TCL) was 391 ± 56 milliseconds (ms).

Nineteen patients (83%) had normal LV systolic function (EF >50%) detected by echocardiogram. All patients’ clinical and demographic characteristics are described in Table 1.

Table 1.

Patients' clinical and demographic characteristics

Patient number	Age (years)	Gender	LV systolic function (EF%)	LA (mm)	Arryhthmia pattern	TCL (msec)	Type of the study	Procedural time (min)	Fluroscopy time (min)	Localization
1	8	Female	56	39	Incessant	380	Conventional	120	90	Nonsept. TA
2	10	Female	62	25	Incessant	410	CARTO 3D	210	45	CS os
3	11	Male	62	35	Paroxysmal	400	Conventional	120	80	CS os
4	8	Male	65	32	Incessant	310	Conventional	120	90	CS os
5	7	Female	56	38	Incessant	360	Conventional	90	15	CS os
6	11	Female	60	32	Paroxysmal	354	CARTO 3D	120	30	Lower CT
7	12	Male	40	41	Incessant	420	CARTO 3D	60	15	Lower FW
8	5	Female	45	40	Incessant	300	Conventional	90	30	CS os
9	10	Female	65	32	Paroxysmal	360	CARTO 3D	120	25	Lower FW
10	11	Male	64	29	Paroxysmal	360	CARTO 3D	120	30	Lower CT
11	8	Male	60	26	Paroxysmal	32o	Conventional	90	40	Nonsept. TA
12	4	Female	55	39	Incessant	300	CARTO 3D	60	10	Nonsept.TA
13	5	Female	56	39	Incessant	350	CARTO 3D	180	25	CS os
14	10	Male	55	29	Incessant	350	Conventional	120	90	CS os
15	10	Female	60	30	Paroxysmal	380	Conventional	90	30	CS os
16	12	Male	63	32	Paroxysmal	380	CARTO 3D	110	25	Lower FW
17	9	Female	56	40	Paroxysmal	380	CARTO 3D	120	30	CS os
18	6	Female	59	39	Incessant	350	CARTO 3D	120	25	CS os
19	8	Female	56	38	Incessant	350	CARTO 3D	90	15	Nonsept. TA
20	10	Male	57	35	Paroxysmal	320	CARTO 3D	120	40	Lower CT
21	11	Female	60	40	Paroxysmal	400	CARTO 3D	120	35	CS os
22	3	Male	50	41	Incessant	360	CARTO 3D	150	40	Nonsept. TA
23	8	Male	40	39	Incessant	380	Conventional	90	45	CS os

LV, left ventricle; EF, ejection fraction; LA, left atrium; TCL, tachycardia cycle length; Nonsept. TA, nonseptal tricuspid annulus; CS os, coronary sinus ostium; Lower CT, crista terminals; Lower FR, lower free wall.

4.2. Mapping and ablation outcomes

Sustained EAT could be induced in 10 patients (44%) during the study using PES or pharmacological provocation, while the rest of patients had incessant EAT (56%) at the beginning of the study as shown in Table 1.

Mapping and ablation were guided by 3D EA mapping using Carto XP system in 61% of patients and by conventional fluoroscopy guided mapping in 39% of patients as shown in Table 1.

Twenty three EAT foci were located in the lower RA and had the following distribution: 12 foci (52%) around coronary sinus ostium (CSo), 5 foci (22%) along nonseptal tricuspid annulus (TA), three foci (13%) were at lower crista terminalis (CT), three foci (13%) at lower free wall (LFW) as shown in Table 1 and Figure 1.

Figure 1.

The anatomical distribution of ectopic atrial tachycardias in the lower right atrium

Successful RF ablation was achieved in all patients with acute success rate of 100%, after a median of 3 (3.3 ± 1.2) RF applications and a median total RF application duration of 4 (4 ± 1.7) minutes. ECG during RF ablation showed acceleration of tachycardia then termination in 55% of cases. The fluoroscopy time was significantly shorter in 3D EA guided mapping compared to conventionally guided mapping (27.8 ± 10 vs 56.7 ± 30 min, respectively, p = 0.003), while there was no significant difference between both mapping techniques in the procedural time (121 ± 40 vs 103 ± 15.8 min, respectively, p = 0.212) as shown in Table 2.

Table 2.

Comparison between 3D EA guided mapping and conventionally guided mapping as regards procedural and fluoroscopic times

Mapping method	Range (ms)	Mean ± SD (ms)	T test
			t	p‐Value
Procedure time
3D EA guided mapping	60–210	121.4 ± 39.9	−1.28	0.212
Conventionally guided mapping	90–120	103.3 ± 15.8
Fluoroscopic time
3D EA guided mapping	10–45	27.8 ± 10.	3.29	0.003
Conventionally guided mapping	15–90	56.6 ± 30.5

4.3. Follow‐up

During the study, four patients had impaired systolic LV function (EF < 50%) which was diagnosed as tachycardia induced cardiomyopathy, following successful RF‐ablation of EAT, systolic LV function was improved during follow up and EF became > 50% by M‐mode detected by transthoracic echocardiography.

4.4. P wave morphology and anatomic location

The area of lower RA is quite a complex anatomical zone with closely packed sites of EAT foci, the anatomical characteristics of this region is detailed in Figure 1 (Table 3).

Table 3.

P wave morphology for lower right atrial ectopic atrial tachycardia

Site	ECG leads
	I	II	III	avR	avF	avL	V1	V2	V3	V4	V5	V6
CSo	−/+ (6) Iso (6)	− (12)	− (12)	+ (12)	− (12)	+ (12)	+ (9) +/− (3)	+ (7) − (5)	− (11) −/+ (1)	− (11) −/+ (1)	− (12)	− (12)
Nonseptal TA	+ (3) Iso (2)	− (5)	− (5)	+ (5)	− (5)	+ (5)	− (5)	− (5)	− (5)	− (5)	− (5)	− (5)
Lower CT	+ (3)	−/+ (2) Iso (1)	−/+ (2) Iso (1)	− (3)	+ (3)	+ (3)	Iso (3)	+ (2) Iso (1)	+ (3)	+ (3)	+ (3)	+ (3)
Lower RA free wall	+ (3)	−/+ (2) − (1)	− (3)	− (3)	− (3)	+ (2) +/− (1)	− (1) + (1) Iso (1)	+ (2) − (1)	− (3)	− (3)	− (3)	− (3)

CSo, coronary sinus ostium; TA, tricuspid annulus; CT, crista terminals; RA, right atrial; ECG, electrocardiogram; iso, isoelectric; +, positive P wave deflection; −, negative P wave; −/+, biphasic P wave with initial negative component; +/−, biphasic P wave with initial positive component.

Electrocardiogram of all study patients with lower RA AT contained a maximum of one positive P wave (up to one lead with exclusively positive pattern) in the inferior (II, III, and aVF) leads; in contrast to EAT originating from the upper RA which typically shows at least two positive P waves in the inferior leads.

4.4.1. Coronary sinus ostium origin

All 12 ATs originating from CSo had a positive pattern in aVL and aVR leads with an exclusively negative pattern in all inferior leads (II, III, and aVF). Eleven of twelve ATs had an exclusively negative pattern in all four precordial leads (V₃–V₆) while only one AT had a biphasic pattern with an early negative component (−/+) in precordial leads V₃ and V₄ (Figure 2).

Figure 2.

P wave morphology was shown on standard 12‐lead electrocardiogram during ectopic atrial tachycardia (EAT) originating from different sites of lower right atrium. TA, tricuspid annulus; CS, coronary sinus; CT, crista terminalis; RA, right atrium

Nine of twelve ATs had positive pattern in lead V₁ while the other three ATs had a biphasic pattern with initial early positive component (+/−) in lead V₁.

4.4.2. Nonseptal TA origin

All five ATs originating at the nonseptal TA (6–9 o'clock) had an exclusively negative pattern in the inferior and precordial (V₃–V₆) leads while they had an exclusively negative pattern in precordial lead V₁ (Figure 2).

4.4.3. Lower crista terminals

All ATs originating from the lower CT (three cases) had an isoelectric pattern in the precordial lead V₁ while they had positive pattern in all four precordial leads (V₃–V₆) (Figure 2).

Two of three ATs had biphasic pattern with an early negative component in inferior leads (II, III, and aVF) while one AT had isoelectric pattern in the inferior leads.

4.4.4. Lower free RA wall

Three ATs originated from the lower free wall of the RA. All exhibited a negative pattern in all precordial leads (V₃–V₆). Two of three ATs had negative pattern in inferior leads (III, aVF) and biphasic pattern (−/+) in lead II while one AT had negative pattern in all inferior leads (II, III, and aVF) (Figure 2).

4.5. Role of precordial leads to discriminate the anatomical atrial tachycardia origin

4.5.1. Lead V₁

In lead V₁, PWM showed a positive pattern during AT originating from CSo (9/12 patients) with a sensitivity of 75% specificity of 91%, positive predictive value (PPV) of 90% and negative predictive value (NPV) of 77% (Table 4). A negative pattern was observed in EAT originating from nonseptal TA (5/5 patients) and lower free wall (1/3 patients), with a sensitivity of 100%, specificity of 94.4%, PPV of 83.3% and NPV of 100% for nonseptal TA‐AT, while for Lower free wall AT sensitivity was 33%, specificity 75%, PPV 17% and NPV 88%. An iso‐electric pattern was observed in the three EAT originating from lower CT with a sensitivity of 100%, specificity of 95%, PPV of 75%, and NPV of 100%.

Table 4.

P wave morphology in precordial leads to discriminate the anatomical lower right atrial tachycardia origin

Anatomic location of EAT	P wave morphology in precordial leads	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)
CSo	+ (V1)	75	91	90	77
	− (V3–V6)	92	27	85	75
Nonseptal TA	− (V1)	100	94.4	83.3	100
	− (V3–V6)	100	22	26	100
Lower CT	Iso (V1)	100	95	75	100
	+ (V3–V6)	100	100	100	100
Lower Free RA wally	− (V1)	33	75	17	88
	− (V3–V6)	100	20	16	100

EAT, ectopic atrial tachycardia; CSo, coronary sinus ostium; TA, tricuspid annulus; CT, crista terminals; RA, right atrial; PPV, positive predictive value; NPV, negative predictive value.

4.5.2. Leads V₃–V₆

In leads V₃–V₆, PWM showed a negative pattern during EAT originating from CSo (12/12), nonseptal TA (5/5) and lower free wall (3/3) with a sensitivity of 92%, 100% and 100%, respectively, specificity of 27%, 22% and 20%, respectively, PPV of 58%, 26% and 16%, respectively, NPV of 75%, 100% and 100%, respectively. In contrast, AT from lower CT showed positive pattern in three of three patients, with an associated sensitivity of 100%, specificity of 100%, PPV of 100% and NPV of 100%.

In this study, the PWM in precordial leads V₁ and V₃–V₆ were not able to discriminate between AT originating from lower free wall and nonseptal TA origin. Both AT origins were associated with predominantly negative P waves in the precordial leads (V₃–V₆).

5. DISCUSSION

In the current series, the low RA was the target site for RF ablation of EAT in 23 out of 31 pediatric patients with right ATs (RATs), representing 74% of the RATs that were successfully ablated in a single institute over a 4 year period. Four distinctive anatomical sites including CSo, nonseptal TA, lower free wall and low CT were the sites of origin of these ATs and had the following distribution 52%, 22%, 13% and 13%, respectively.

Though ATs arising from low RA represented the majority of RATs in the present series, other reports described it as an uncommon AT accounting for <20% of RATs (Huo et al., 2013; Kistler et al., 2006). The apparent discrepancy may be owed to the small sample size of the current study and restriction of the study to pediatric age group compared to other reports. In Kistler et al.'s (2006) work, the lower RA foci were more commonly located at the TA (n = 38) followed by CSo (n = 16) and in Huo et al. (2013) work which exclusively included adult population, the most common sites were TA and CSo. In contrast to our findings where CSo foci were the most commonly encountered in pediatric patients (52%).

Several studies have described the different anatomical sites of EAT arising from the low RA region detailing their electrocardiographic and electrophysiological features, separately (Cosio et al., 1999; Kistler et al., 2006; Morton et al., 2001). Anatomically these sites lie in close proximity to each other based on the structural properties of Low RA. Consequently, the currently available ECG algorithms are somehow deficient to distinguish one from the other.

5.1. P wave morphology in precordial leads

In this study, PWM in precordial leads has been able to localize and distinguish between different anatomical sites of EAT in low RA in pediatric age group with high sensitivity and specificity in the majority of sites.

P wave morphology in leads V₃–V₆ has been able to localize EAT arising from low CT with a high PPV and NPV while in low TA (CSo+ nonseptal TA) or LFW locations, it provided high NPV but a poor PPV, in contrast to Huo et al. (2013) work which was conducted on adult population the leads V₃–V₆ has been able to provide high PPVs and NPVs for the localization of AT originating from the lower TA or CT.

On the other hand, in this study PWM in lead V₁ has been able to localize EAT originating from low TA or low CT with high PPV and NPV, which was not shown in other reports.

The Power of precordial leads to discriminate between different lower RA sites.

P wave morphology in leads V₃–V₆ showed distinctive patterns discriminating EATs arising from the area of low TA (CSo+ nonseptal TA) and lower free wall from those originating from low CT.

Showing a predominantly negative PWM or an early negative component in AT originating from CSo and exclusively negative PWM in nonseptal TA and LFW ATs in contrast to exclusively positive PWM in lower CT ATs, these patterns were consistent with patterns reported by other studies (Huo et al., 2013; Kistler et al., 2006; Morton et al., 2001). However, using PWM in leads V₃–V₆ alone failed to differentiate between EATs arising from the adjacent locations of CSo, nonseptal TA and LFW.

In our study, A positive PWM in lead V₁ has been able to discriminate CSo origin from nonseptal TA and LFW location with a high sensitivity (75%) and specificity (91%) which Huo et al. (2013) study failed to demonstrate.

6. CONCLUSION

Adding the PWM in precordial leads to the currently available ECG based algorithms can help differentiate between most EAT originating from different locations in low RA in pediatric age group.

7. STUDY LIMITATIONS

The small sample size which was an inevitable limitations due to restriction of the study to a specific age group as well as the inherit low prevalence of EATs originating from low RA compared to other atrial locations.

DISCLOSURES

Dr. Lamyaa Allam, Dr. Rania Samir, and Dr. Mazen Tawfik declare that they have no conflicts of interest.

COMPLIANCE WITH ETHICAL STANDARDS

This study was approved by our ethical committee and an informed consent was obtained from all patients and their parents prior to their inclusion in the study.

Allam LE, Ahmed RS, Ghanem MT. Role of surface electrocardiogram precordial leads in localizing different anatomic sites of ectopic atrial tachycardia arising from lower right atrium in pediatric population. Annals of Noninvasive Electrocardiol. 2018;23:e12485 10.1111/anec.12485