Skip to main content
NCBI home page
Annals of Noninvasive Electrocardiology logoLink to Annals of Noninvasive Electrocardiology
. 2017 Dec 18;23(4):e12525. doi: 10.1111/anec.12525

Relation between electrical and mechanical dyssynchrony in patients with left bundle branch block: An electro‐ and vectorcardiographic study

Jan De Pooter 1,✉, Milad El Haddad 1, Victor Kamoen 1, Thomas Tibin Kallupurackal 1, Roland Stroobandt 1, Marc De Buyzere 1, Frank Timmermans 1
PMCID: PMC6931644  PMID: 29251398

Abstract

Background

Current guidelines select patients for cardiac resynchronization therapy (CRT) mainly on electrocardiographic parameters like QRS duration and left bundle branch block (LBBB). However, among those LBBB patients, heterogeneity in mechanical dyssynchrony occurs and might be a reason for nonresponse to CRT. This study assesses the relation between electrocardiographic characteristics and presence of mechanical dyssynchrony among LBBB patients.

Methods

The study included patients with true LBBB (including mid‐QRS notching) on standard 12‐lead electrocardiograms. Left bundle branch block‐induced mechanical dyssynchrony was assessed by the presence of septal flash on two‐dimensional echocardiography. Previously reported electro‐ and vectorcardiographic dyssynchrony markers were analyzed: global QRS duration (QRSDLBBB), left ventricular activation time (QRSDLVAT), time to intrinsicoid deflection (QRSDID), and vectorcardiographic QRS areas in the 3D vector loop (QRSA 3D).

Results

The study enrolled 545 LBBB patients. Septal flash (SF) is present in 52% of patients presenting with true LBBB. Patients with SF are more frequent female, have less ischemic heart disease and smaller left ventricular dimensions. In multivariate analysis longer QRSDLBBB, QRSDLVAT and larger QRSA 3D were independently associated with SF. Of all parameters, QRSA 3D has the best accuracy to predict SF, although overall accuracy remains moderate (59% sensitivity, 58% specificity). The predictive value of QRSA 3D remained constant in both sexes, irrespective of ischemic heart disease, ejection fraction and even when categorizing for QRSDLBBB.

Conclusion

In LBBB patients, large QRS areas correlate better with mechanical dyssynchrony compared to wide QRSD intervals. However, the overall accuracy to predict mechanical dyssynchrony by electrocardiographic dyssynchrony markers, even when using complex vectorcardiographic parameters, remains low.

Keywords: cardiac resynchronization therapy, dyssynchrony, electrocardiography, vectorcardiography and left bundle branch block

1. INTRODUCTION

Current guidelines on cardiac resynchronization therapy (CRT) select patients mainly on electrocardiographic criteria such as QRS duration (QRSD) and QRS morphology (Ponikowski et al., 2016; Yancy et al., 2013). These criteria refer to the electrical dyssynchrony caused by block of the left bundle branch (LBBB) as the substrate for CRT. However, it has been shown that patients with LBBB morphology and wide QRSD reveal variable ventricular activation patterns. This heterogeneity in mechanical dyssynchrony among patients with LBBB, is thought to be one of the reasons why a significant number of patients fail to respond to CRT (Auricchio et al., 2004). Several new electro‐ and vectorcardiographic parameters have been proposed as markers of both electrical and mechanical dyssynchrony (Del‐Carpio Munoz et al., 2013; van Deursen et al., 2015; Mafi Rad et al., 2016; Sweeney et al., 2010). However, these parameters were validated against different dyssynchrony assessments and not compared head to head. Recently, a simple visual assessment of LBBB‐induced mechanical dyssynchrony, called SF, has been introduced. This SF refers to an early rapid inward motion of the septum on echocardiography and has been shown to be a strong and independent predictor of CRT response (Stankovic et al., 2016, 2017). Moreover, visual assessment of SF is an accurate, highly reproducible and easy parameter to diagnose mechanical dyssynchrony. This study aims to assess (i) the prevalence and determinants of SF among patients with true LBBB and (ii) whether new electro‐ and vectorcardiographic dyssynchrony parameters correlate with the presence of SF.

2. METHODS

2.1. Study design and selection of patients

The study enrolled patients with true LBBB morphology on a standard 12‐lead electrocardiogram (ECG) at the Cardiologic department of the University Hospital of Ghent between June 2013 and September 2016. True LBBB was defined according to the recent American Heart Association, American College of Cardiology Foundation and Heart Rhythm Society criteria including: QRSD ≥ 120 ms, QS or rS in lead V1 and broad notched or slurred R waves in two adjacent leads among leads I, aVL, V5 and V6(Surawicz et al., 2009). The presence of mid‐QRS notching and slurring in the left lateral leads was included as this characteristic differentiates true LBBB from QRS prolongation due to left ventricular hypertrophy (Strauss & Selvester, 2009; Strauss, Selvester, & Wagner, 2011). All ECGs were recorded with MAC 5500 ECG recording devices (GE Healthcare, Waukesha, WI, USA) and stored digitally (aHL7 ECGs, sampling rates of 500 Hz) in a MUSE Cardiology Information system (GE Healthcare). The study was approved by the ethical committee of the University Hospital of Ghent.

2.2. Electrocardiographic parameters to assess dyssynchrony

QRSD intervals were measured automatically using the Marquette 12SL algorithm. Left bundle branch block QRSD (QRSDLBBB) is measured as a global QRSD, which is calculated from the earliest beginning until the latest ending of the QRS complex in all leads, as recommended by guidelines (Surawicz et al., 2009) (Figure 1). This automated algorithm was previously validated in LBBB patients by comparing it to manual QRSD measurements using digital calipers (De Pooter, El Haddad, Stroobandt, De Buyzere, & Timmermans, 2017b). Besides QRSDLBBB, two other QRSD intervals, which have been proposed as makers of both electrical and mechanical dyssynchrony, were calculated (Figure 1). QRSDLVAT is defined as the interval from the first notch to the end of the QRS complex and represents the delayed activation time of the left ventricle in LBBB patients (Sweeney et al., 2010). QRSDID represents the time from the earliest onset of the QRS complex to the latest peak or point at which the maximum deflection (intrinsicoid deflection) to baseline occurs. In LBBB patients this QRSDID is maximal in the left lateral leads and therefore proposed as marker of delayed left ventricular activation and dyssynchrony (Del‐Carpio Munoz et al., 2013).

Figure 1.

Figure 1

Open in a new tab

Electro‐ and vectorcardiographic measurements in left bundle branch block patients to assess both electrical and mechanical dyssynchrony

2.3. Vectorcardiographic parameters to assess dyssynchrony

Custom‐made software (Matlab software, Mathworks, Natick, MA, US) was used to convert digital ECGs to vectorcardiograms (VCG) according to Frank's orthogonal lead system as previously reported (De Pooter et al., 2017a). Each VCG was plotted against the three orthogonal leads (X, Y, and Z) allowing to form a 3D vector. QRS areas (QRSA) are calculated as the integral between the QRS waveform and baseline in each orthogonal lead (QRSAX, QRSAY, and QRSAZ) (Figure 1). The QRS area of the 3D vector loop (QRSA3D) was calculated as (QRSAX + QRSAY 2 + QRSAZ 2)½ and has been previously validated as a marker of ventricular dyssynchrony (van Deursen et al., 2015). We recently showed that LBBB patients are characterized by large QRS areas in the Z‐lead. Therefore, QRSAZ was evaluated separately (De Pooter et al., 2017a). Additionally, as QRS areas in individual leads of the standard 12‐lead ECG have not been investigated, we calculated the QRS area in each lead separately (QRSAI, II, II, aVL, aVR, aVF, V1, V2, V3, V4, V5, and V6).

2.4. Echocardiographic studies and assessments of mechanical dyssynchrony

Echocardiographic examinations within 3 months of the ECG recording date were considered for analysis. All echocardiographic examinations were performed using commercially available systems (GE Healthcare Ultrasound Vivid 7 and GE Healthcare Ultrasound Vivid E9, Vingmed, Horton, Norway; Philips Ultrasound iE 33, Best, The Netherlands). Two echocardiography experts, blinded to the ECGs, reviewed all echocardiographic studies offline using EchoPAC version 7.1.13 for the GE scanning systems and Xcelera viewer R3 version 3.3.1 2013 for the Philips scanning system.

Left ventricular (LV) dimensions were measured in conventional parasternal views using LV enddiastolic diameter (LVEDD), LV endsystolic diameter (LVESD), interventricular septal wall thickness (IVSD), and posterior wall thickness (PWD). Left ventricular mass (LVMASS) was calculated as LVMASS (g) = 0.8 (1.04 ([LVEDD + IVSD + PWD]3−LVEDD 3)) + 0.6 (Lang et al., 2015). Left ventricular EDD, LVESD and LVMASS were indexed for body surface area (BSA): LVEDDi, LVESDi and LVMASSi. Ejection fraction (EF) was graded according to: normal (EF > 55%), slightly depressed (EF 55%–45%), moderately depressed (EF 44%–35%), severely depressed (EF < 35%).

Mechanical dyssynchrony was assessed by the presence of septal flash (SF) on two‐dimensional echocardiography. SF refers to a specific echocardiographic pattern in which a rapid, pre‐ejection, leftward motion (right to left) of the septum occurs. The presence of SF was assessed visually (parasternal short axis, parasternal long axis or apical views) as validated in prior studies and at our center (Corteville et al., 2017; Stankovic et al., 2016, 2017).

2.5. Statistical analysis

Continuous variables are expressed as mean ± SD or median [quartile 1; quartile 3] if data were not Gaussian distributed. Categorical variables are expressed as absolute numbers with percentage (%). Shapiro‐Wilk test was used to test for normality. Univariate comparison among groups was done by Mann–Whitney U test and categorical variables were compared by Chi‐square tests. Significant variables in univariate analysis were subsequently tested in a multivariate analysis using multiple linear regression. Non‐Gaussian distributed variables were log transformed. Multicollinearity was defined as a variance inflation factor >4. Correlations between continuous variables were analyzed using Spearman rank‐order correlation coefficients. Receiver operating characteristic (ROC) curves were constructed to compare the ability of electro‐ and vectorcardiographic parameters in predicting SF. Statistical significance was set at a two‐tailed probability level of < 0.05. All statistical analysis was performed using SPSS software (version 24.0, IBM, Armonk, NY, USA).

3. RESULTS

3.1. Patient characteristics and prevalence of septal flash

The study enrolled 605 LBBB patients. In 60 patients, assessment of SF was not possible due to inappropriate image quality. Therefore, 545 LBBB patients were considered for further analysis. The cohort compromised 217 (40%) females and mean age was 74 ± 15 years. Ischemic heart disease was prevalent among 230 (42%) of the patients. Patient characteristics are summarized in Table 1.

Table 1.

Patient characteristics

Patient group All patients (n = 545)
Prevalence of septal flash n (%) 284 (52)
Baseline characteristics
Age (years) 74 ± 15
Length (cm) 168 ± 10
Weight (kg) 74 ± 16
Body mass index (kg/m²) 26 ± 5.1
Body surface area (BSA; m²) 1.84 ± 0.22
Blood pressure systolic (mmHg) 130 ± 36
Blood pressure diastolic (mmHg) 67 ± 19
Heart rate (beats/min) 71 ± 20
Underlying heart disease n (%)
Ischemic heart disease 230 (42)
Congenital heart disease 21 (3.9)
Valvular heart disease 132 (24)
Atrial fibrillation 48 (8.8)
Echocardiographic measurements
Enddiastolic diameter (mm) 50 [45;56]
Enddiastolic diameter/BSA (mm/m²) 27 [25;30]
Endsystolic diameter (mm) 33 [27;41]
Endsystolic diameter/BSA (mm/m²) 18 [15;22]
Left ventricular mass (g) 198 [156;249]
Left ventricular mass/BSA (g/m²) 108 [87;131]
Ejection fraction
Normal (>55%) 262 (48)
Mildly reduced (45%–54%) 124 (23)
Moderately reduced (35%–44%) 95 (17)
Severely reduced (<35%) 64 (12)
Open in a new tab

3.2. Electro‐ and vectorcardiographic measurements in LBBB patients

Electro‐ and vectorcardiographic measurements are summarized in Table 2. Mean QRSDLBBB was 148 [140;162] ms for all patients. Females had smaller QSRDLBBB (144 [136;153] ms) compared to males (154 [142;168]ms, p < .001). Similar differences (± 10 ms) between males and females were found for QRSDLVAT and QRSDID. No differences were found in QRSDLBBB, QRSDLVAT, and QRSDID intervals between ischemic and nonischemic patients (Table 2).

Table 2.

ECG and VCG measurements

ECG‐VCG parameter All Patients (n = 545) Females (n = 217) Males (n = 328) p ‐value Ischemic (n = 230) Nonischemic (n = 315) p‐value
QRSDLBBB (ms) 148 [140;162] 144 [136;153] 154 [142;168] <.001 150 [140;164] 148 [140;162] .189
QRSDID (ms) 68 [56;86] 62 [53;73] 72 [60;92] <.001 68 [58;88] 66 [56;86] .235
QRSDLVAT (ms) 91 [83;104] 87 [79;95] 96 [84;110] <.001 93 [83;106] 91 [83;104] .405
QRSA3D (μVs) 117 [93;147] 112 [88;132] 122 [98,156] <.001 112 [88;137] 125 [98;156] <.001
QRSAZ (μVs) 88 [59;122] 83 [59;107] 93 [59;132] .005 78 [54;112] 98 [63;127] .001
Open in a new tab

ECG, electrocardiographic; VCG, vectorcardiographic; QRSDLBBB, global QRS duration; QRSDID, time to intrinsicoid deflection; QRSDLVAT, left ventricular activation time; QRSA3D, vectorcardiographic calculated QRS area in the 3D vector loop; QRSAZ, vectorcardiographic calculated QRS area in the Z lead.

Overall mean QRSA3D was 117 [93;147]μVs, with higher QRSA3D in males compared to females (122 [98;156] μVs vs. 112 [88;132]μVs, p < .001). Nonischemic patients showed higher QRSA3D compared to ischemic patients (125 [98;156]μVs vs. 112 [88:137] μVs, p < .001). QRSAZ measurements showed similar trends among these patient groups.

Correlation between QRSDLVAT and QRSDLBBB was higher (r s: 0.80, p < .001) compared to the correlation between QSRDID and QRSDLBBB (r s, 0.43 p < .001). QRSA3D and QRSAZ had moderate but significant correlation with QRSDLBBB (r s: 0.40, p < .001 and r s: 0.30, p < .001 respectively).

3.3. Clinical, echo‐, electro‐, and vectorcardiographic determinants of patients with septal flash

Compared to patients without SF, patients with SF were more frequent female (46% vs. 33%, p = .002), had less ischemic heart disease (35% vs. 50%, p = .001) and had smaller LV dimensions measured by LVMASS (187 [153;241] g vs. 205 [164;261]g, p = .017). QRSDLBBB, QRSDLVAT, QRSA3D, and QRSAZ were higher in patients with SF (Table 3). A trend toward more patients with severely depressed LV function (<35%) could be observed in the group of patients with SF vs. those without SF (10 vs. 14%, p = .091).

Table 3.

Determinants of septal flash, univariate and multivariate analysis

Septal flash No septal flash p‐value
n = 284 n = 261 Univariate Multivariate
Clinical characteristics
Age (yrs) 69 ± 14 71 ± 15 .066
Female n (%) 131 (46) 86 (33) .002 .021
Length (cm) 167 ± 10 167 ± 10 .406
Weight (kg) 75 ± 15 76 ± 17 .849
Body mass index (kg/m²) 27 ± 5 27 ± 5 .826
Body surface area (BSA) (m²) 1.83 ± 0.21 1.85 ± 0.23 .603
Heart rate (beats/min) 74 ± 19 75 ± 22 .790
Underlying heart disease
Ischemic heart disease n (%) 99 (35) 131 (50) .001 .004
Congenital heart disease 12 (4.2) 9 (3.5) .659
Echocardiographic measurements
End‐diastolic diameter (mm) 49 [45;49] 51 [45;56] .085
End‐diastolic diameter/BSA (mm/m²) 27 [25;29] 27 [34;30] .995
End‐systolic diameter (mm) 33 [27;40] 33 [27;42] .686
End‐systolic diameter/BSA (mm/m²) 18 [15;21] 18 [14;22] .989
Left ventricular mass (g) 187 [153;241] 205 [164;261] .017 .011
Left ventricular mass/BSA (g/m²) 101 [86;127] 114 [89;137] .060
Ejection fraction
Normal (>55%) 142 (50) 120 (46) .347
Mildly reduced (45–54%) 69 (24) 55 (21) .370
Moderately reduced (35–44%) 46 (16) 49 (19) .428
Severely reduced (<35%) 27 (10) 37 (14) .091
ECG and VCG measurements
QRSDLBBB (ms) 150 [140:164] 146 [136;160] .026 .007
QRSDID (ms) 68 [56;86] 68 [56:84] .882
QRSDLVAT (ms) 93 [83;106] 89 [79;102] .026 .046
QRSA3D (μVs) 125 [103;156] 107 [85;137]  < .001  < .001
QRSAZ (μVs) 98 [68;127] 78 [49;112]  < .001  < .001
Open in a new tab

Bold values indicate significance (p < .05). ECG, electrocardiographic; VCG, vectorcardiographic; QRSDLBBB, global QRS duration; QRSDID, time to intrinsicoid deflection; QRSDLVAT, left ventricular activation time; QRSA3D, vectorcardiographic calculated QRS area in the 3D vector loop; QRSAZ, vectorcardiographic calculated QRS area in the Z lead.

In a multiple regression model including gender, ischemic heart disease and LVMASS, each of QRSDLBBB, QRSDLVAT, QRSA3D, and QRSAZ was independently associated with higher SF prevalence (Table 3).

Receiver operating characteristic curves to predict the presence of SF by electro‐ and vectorcardiographic dyssynchrony markers showed areas under the curve (AUC) ranging from 0.519 to 0.674 (Table 4). Of all parameters, QRSA3D (AUC: 0.674) and QRSAZ (AUC: 0.661) revealed the best accuracy to detect SF among LBBB patients and performed significantly better compared to QRSDLBBB (AUC: 0.587, p = .03 and p = .04 respectively). The ROC curve for QRSA3D revealed an optimal cutoff at 114μVs, showing a sensitivity of 59% with 58% specificity to predict SF by QRSA3D. The diagnostic accuracy of QRSDLVAT (AUC: 0.621) did not outperform QRSDLBBB (p = .600) in predicting SF. QRSID revealed the lowest AUC: (0.519) to discriminate LBBB patients with SF from those without SF.

Table 4.

Diagnostic value of ECG and VCG parameters in assessing the presence of septal flash

ECG‐VCG parameter Areas under the curve to predict septal flash
All (n = 545) Females (n = 217) Males (n = 328) Ischemic (n = 230) Nonischemic (n = 315) QRSD <130 QRSD 130–149 QRSD >150
QRSDLBBB 0.555 0.473 0.644 0.540 0.594 0.535 0.582 0.515
QRSDID 0.519 0.504 0.529 0.550 0.502 0.578 0.495 0.452
QRSDLVAT 0.621 0.559 0.653 0.627 0.620 0.638 0.646 0.516
QRSA3D 0.674 0.712 0.668 0.660 0.672 0.647 0.644 0.657
QRSAZ 0.661 0.685 0.660 0.637 0.684 0.631 0.632 0.653
Open in a new tab

ECG, electrocardiographic; VCG, vectorcardiographic; QRSDLBBB, global QRS duration; QRSDID, time to intrinsicoid deflection; QRSDLVAT, left ventricular activation time; QRSA3D, vectorcardiographic calculated QRS area in the 3D vector loop; QRSAZ, vectorcardiographic calculated QRS area in the Z lead. QRS durations are measured in ms, QSRA are measured in μVs.

Moreover, the accuracy of QRSDLBBB, QRSDLVAT and QRSDID varied strongly among different patients groups (Table 4). In male LBBB patients, QRSD intervals showed higher accuracy to diagnose mechanical dyssynchrony compared to females. On the other hand, the diagnostic accuracy of QRSA3D and QRSAZ remained constant among different patient groups based on sex or ischemic vs. nonischemic heart disease (range AUCs QRSA3D: 0.660–0.712 and range AUCs QRSAZ 0.633–0.685). Of interest, even when categorizing for QRSDLBBB, QRSA3D and QRSAZ showed stable AUCs over the entire range of QRSD with AUC ranging from 0.631–0.657 (Table 4).

3.4. Relation between electrical and mechanical dyssynchrony according to ejection fraction

A subanalysis was conducted for patients meeting a CRT indication (EF < 35% and clinical heart failure, n = 64) vs. those without current CRT indication (EF ≥ 35%, n = 481). Patients with depressed EF (<35%) had wider QRSDLBBB and QRSDLVAT, whereas QRSDID, QRSA3D and QRSAZ were not significantly different between the patients groups. Of all parameters, QRSA3D had the highest AUC to predict septal flash, both in patients with (AUC: 0.620) or without CRT indication (AUC: 0.684). QRSDLBBB and QRSID had the lowest AUC to predict SF in both groups. All ECG‐VCG parameters showed slightly lower AUCs (p = not significant) for prediction of SF in patients with EF < 35%. (Table 5).

Table 5.

ECG and VCG values and the correlation with septal flash, stratified for left ventricular ejection fraction

ECG‐VCG parameter AUC to predict SF
EF ≥ 35% (n = 481) EF<35% (n = 64) p‐value EF ≥ 35% (n = 481) EF<35% (n = 64) p‐value
QRSDLBBB 148 (138;152) 158 [146;172] <.001 0.588 0.567 .797
QRSDID 66 (56;86) 72 [61;86] .100 0.533 0.462 .387
QRSDLVAT 91 (81;104) 100 [89;113] <.001 0.639 0.567 .377
QRSA3D 117 (93;147) 117 [93;150] .999 0.684 0.620 .422
QRSAZ 88 (59;122) 90 [54;122] .601 0.673 0.591 .310
Open in a new tab

Bold values indicate significance (p < .05). ECG, electrocardiographic; VCG, vectorcardiographic; AUC, area under the curve; EF, left ventricular ejection fraction; SF, Septal flash; QRSDLBBB, global QRS duration; QRSDID, time to intrinsicoid deflection; QRSDLVAT, left ventricular activation time; QRSA3D, vectorcardiographic calculated QRS area in the 3D vector loop; QRSAZ, vectorcardiographic calculated QRS area in the Z lead. QRS durations are measured in ms, QSRA are measured in μVs.

3.5. QRS areas in individual leads of the VCG and ECG

QRSA3D correlated strongly with QRSAZ (r s: 0.875, p < .001), whereas QRSAX (r s: 0.115, p < .007) and QRSAY (r s: 0.340, p < .001) showed poor correlation with QRSA3D. Of all leads of the standard 12‐lead ECG, lead QRSAV1 and QRSAV2 showed the best correlation with QRSA3D (r s: −0.744, p < .001 and r s: −0.827, p < .001 respectively). As such QRSAV1 and QRSAV2 revealed the highest accuracy to diagnose SF (AUC: 0.665 and 0.637, Figure 2). Correlation of left lateral leads with QRSA3D was low (r s all <0.500 for QRSAI, QRSAaVL, QRSAV5 and QRSAV6).

Figure 2.

Figure 2

Open in a new tab

Receiver operating characteristics curves to assess the presence of septal flash by electro‐ and vectorcardiographic parameters. Of all parameters, QRS area in the 3D vector loop (QRSA 3D) and QRSA area in the Z lead of Frank's vectorcardiogram (QRSAZ) revealed the highest accuracy. As the QRS areas calculated in lead V1 (QRSAV 1) and lead V2 (QRSAV 2) of the standard 12‐lead electrocardiogram correlate highly with QRSA 3D and QRSAZ, areas in leads V1 and V2 show similar diagnostic accuracy compared to vectorcardiograms calculated QRS areas

4. DISCUSSION

4.1. Mechanical dyssynchrony in patients with LBBB

Patients with LBBB benefit more from CRT than patients with non‐LBBB (Sipahi et al., 2012). This is explained as patients with LBBB have a desynchronized ventricular contraction caused by delayed activation of the lateral left ventricle, which is most likely to be resynchronized with CRT. However, even when applying current CRT selection criteria, some patients with LBBB and wide QRSD fail to respond to CRT. One of the reasons could be that not all LBBB patients display the same mechanical activation pattern which can be corrected by CRT. For instance, Auricchio et al. (2004) showed that among patients with LBBB, heterogeneity in ventricular activation patterns exists. The current study shows that when applying strict criteria for diagnosing LBBB (including mid‐QRS notching), SF is merely present in 52% of the patients, indicating that the presence of electrical dyssynchrony (as defined by wide QRSD and LBBB) does not coincide completely with the presence of LBBB induced mechanical dyssynchrony (as assessed by SF). The presence of SF might therefore identify a particular subset of LBBB patients.

Electrophysiological studies in patients with SF revealed long transseptal activation times, attributed to slow muscle to muscle conduction in the septum (Duckett et al., 2012). Auricchio et al. (2004) showed that the majority of LBBB patients in his study had long transseptal activation times and revealed a typical U‐shaped activation pattern. However, one‐third of the patients with LBBB did not show this activation pattern and several early breakthrough sites in the septum occurred leading to shorter transseptal activation times. Although no SF assessments were performed in that study, we hypothesize that those patients with septal breakthroughs are those LBBB patients without SF. In an experimental study by Gjesdal et al. (2011), radiofrequency ablation of the proximal part of the left bundle in dogs results in LBBB with typical characteristics of SF. These LBBB‐induced dog hearts eventually developed LV dysfunction, which could be restored with CRT. This suggests that SF associated with typical LBBB, is probably caused by proximal block of the left bundle branch in humans (Lumens et al., 2013).

4.2. Correlation of mechanical dyssynchrony with electro‐ and vectorcardiographic parameters

In patients with conduction disorders, delayed activation of the ventricle is reflected as wide QRSD on the ECG. In the field of CRT, wide QRSD is used as a marker for electrical dyssynchrony in patients with heart failure. Given the heterogeneity and disparity between QRSD and mechanical dyssynchrony, several new electro‐ and vectorcardiographic parameters have been developed which claim to reflect both electrical and mechanical dyssynchrony. These parameters use either a well‐defined part of the QRSD interval or calculate the surface of the QRS waveform (QRS area) (Del‐Carpio Munoz et al., 2013; van Deursen et al., 2015; Mafi Rad et al., 2016; Sweeney et al., 2010). Although these parameters have shown to better reflect mechanical dyssynchrony and CRT outcome compared to QRSD in different studies, no study compared those dyssynchrony markers head‐to‐head and against SF. Our study compared these novel markers in a well‐defined population of patients with true LBBB and using SF as a valid and reproducible measure of mechanical dyssynchrony.

Of all parameters, VCG‐calculated QRS areas correlated best with SF. This is in line with a previous study, which showed that large QRS areas are associated with a higher degree of mechanical dyssynchrony, measured by electro‐anatomic mapping (Mafi Rad et al., 2016). Of interest, the accuracy of QRS areas to assess mechanical dyssynchrony is robust over different patient groups based on gender, presence or absence of ischemic heart disease and ranges of QRSD. Conversely, QRSD parameters seem to mainly be correlated with mechanical dyssynchrony in males and not in females. This is most probably explained as though SF is more prevalent in females, it occurs more frequent at narrower QRSD compared to males.

Although the number of patients requiring CRT in our study population was limited, our results indicate that the relation between electrical and mechanical dyssynchrony might be different for patients with EF < 35% vs. those with EF ≥ 35%. Most of the studied QRS characteristics indeed showed lower AUC to predict SF by ECG‐VCG parameters in patients with depressed EF, although QRSA3D remained the best predictor of SF both for patients with EF < 35% and ≥35%.

Interestingly, in patients with narrow QRSDLBBB (<150 ms) specific QRSD intervals, like QRSDLVAT, correlate better with mechanical dyssynchrony compared to overall QRSDLBBB. This might indicate that additional QRS measurements, like QRSDLVAT, might be of value as they might specify a particular subset of patients with narrow QRSDLBBB and mechanical dyssynchrony.

Vectorcardiograms‐calculated QRS area combines both the information of the QRS morphology and duration into one single parameter. Patients with LBBB have QRS areas 2–3 times larger compared to patients without conduction delay (van Deursen et al., 2015). These large 3D QRS areas in LBBB patients can be explained by strong unopposed electrical forces generated by delayed activation of the posterior and basal parts of the LV, typically seen in LBBB (Ploux et al., 2013). The largest QRS areas are therefore detected in antero‐posterior oriented leads, such as the Z‐lead of the Franks orthogonal system or lead V1 and V2 of the 12‐lead ECG. Hence, QRS areas in the Z, V1 or V2 lead are highly correlated with 3D QRS areas in LBBB patients. Therefore, our results show that measuring 3D QRS areas in LBBB patients can be simplified to calculations of QRS areas in lead V1 or V2 of the standard 12‐lead ECG.

5. CLINICAL IMPLICATIONS

Current CRT guidelines select patients by QRSD cutoffs and QRS morphology (Ponikowski et al., 2016; Yancy et al., 2013). However, with current selection criteria, up to one‐third of the patients do not achieve the expected CRT response (Auricchio & Prinzen, 2011). This number of nonresponders, despite these patients meet the current selection criteria, might largely be attributed to disparity between electrical and mechanical dyssynchrony. In the last years, emerging evidence exists that the presence of SF in LBBB patients is an important determinant of long term CRT response, with an incremental value over clinical variables and QRSD (Stankovic et al., 2016, 2017). The PREDICT‐CRT trial, which included 1060 CRT patients, showed that the presence of SF and its correction with CRT predicts both long‐term reverse remodeling and all‐cause mortality. Additionally, multiparametric scoring models to select heart failure patients for CRT treatment came to the same conclusion (Brunet‐Bernard et al., 2014; Maass et al., 2017). In these models, the inclusion of simple visual assessments of mechanical dyssynchrony, like SF, identified better CRT responders compared to score models without assessing SF. These models therefore consider LBBB‐morphology, QRS area and presence of SF as independent predictors of CRT response. Likewise, we showed that ECG‐derived parameters (even complex VCG‐calculated parameters) cannot identify with high accuracy those LBBB patients with SF from those without SF. Even with the best parameter (QRSA3D), sensitivity and specificity do not reach 60%. Therefore, SF might be suggested as an additional marker, independently or on top of ECG characteristics, of those LBBB patients who will likely respond to CRT.

6. LIMITATIONS

This study was conducted as a retrospective study. Mechanical dyssynchrony was assessed solely by the presence of SF, and therefore no conclusions can be drawn with respect to other markers of inter‐ or intraventricular dyssynchrony. Patients were selected on LBBB morphology as defined by the American Heart Association, American College of Cardiology Foundation and Heart Rhythm Society. This definition includes the presence of mid‐QRS notching. Other LBBB definitions may select other patients with other clinical and echocardiographic characteristics and therefore yield a different prevalence of SF. The majority of patients did not have heart failure or reduced EF. As such, our results should be interpreted with caution, as heart failure patients with LBBB might differ from our population. Our study did not assess outcome, whether SF predicts clinical outcome in a more general LBBB population mandates future studies.

7. CONCLUSION

Mechanical dyssynchrony, as assessed by SF, is present in half of the patients presenting with true LBBB on the ECG. Among these patients mechanical dyssynchrony correlates better with larger QRS areas compared to wider QRSD intervals. However, the overall accuracy to predict mechanical dyssynchrony by electrocardiographic dyssynchrony markers, even when using complex vectorcardiographic parameters, remains rather low. Our study emphasizes that the selection of CRT candidates solely by electrocardiographic parameters might therefore not necessarily select those patients with the best electro‐mechanical substrate for CRT and that mechanical and electrical dyssynchrony should preferably assessed by separate parameters.

DISCLOSURES

None.

Pooter JD, Haddad ME, Kamoen V, et al. Relation between electrical and mechanical dyssynchrony in patients with left bundle branch block: An electro‐ and vectorcardiographic study. Ann Noninvasive Electrocardiol. 2018;23:e12525 10.1111/anec.12525

REFERENCES

  1. Auricchio, A. , Fantoni, C. , Regoli, F. , Carbucicchio, C. , Goette, A. , Geller, C. , … Klein, H. (2004). Characterization of left ventricular activation in patients with heart failure and left bundle‐branch block. Circulation, 109(9), 1133–1139. 10.1161/01.CIR.0000118502.91105.F6 [DOI] [PubMed] [Google Scholar]
  2. Auricchio, A. , & Prinzen, F. W. (2011). Non‐responders to cardiac resynchronization therapy: The magnitude of the problem and the issues. Circulation Journal, 75(3), 521–527. 10.1253/circj.CJ-10-1268 [DOI] [PubMed] [Google Scholar]
  3. Brunet‐Bernard, A. , Marechaux, S. , Fauchier, L. , Guiot, A. , Fournet, M. , Reynaud, A. , & Donal, E. (2014). Combined score using clinical, electrocardiographic, and echocardiographic parameters to predict left ventricular remodeling in patients having had cardiac resynchronization therapy six months earlier. American Journal of Cardiology, 113(12), 2045–2051. 10.1016/j.amjcard.2014.03.050 [DOI] [PubMed] [Google Scholar]
  4. Corteville, B. , De Pooter, J. , De Backer, T. , El Haddad, M. , Stroobandt, R. , & Timmermans, F. (2017). The electrocardiographic characteristics of septal flash in patients with left bundle branch block. Europace, 19, 103–109. [DOI] [PubMed] [Google Scholar]
  5. De Pooter, J. , El Haddad, M. , De Buyzere, M. , Aranda, H. A. , Cornelussen, R. , Stegemann, B. , … Timmermans, F. (2017. a). Biventricular Paced QRS Area Predicts Acute Hemodynamic CRT Response Better than QRS Duration or QRS Amplitudes. Journal of Cardiovascular Electrophysiology, 28(2), 192–200. 10.1111/jce.13132 [DOI] [PubMed] [Google Scholar]
  6. De Pooter, J. , El Haddad, M. , Stroobandt, R. , De Buyzere, M. , & Timmermans, F. (2017. b). Accuracy of computer‐calculated and manual QRS duration assessments: Clinical implications to select candidates for cardiac resynchronization therapy. International Journal of Cardiology, 236, 276–282. 10.1016/j.ijcard.2017.01.129 [DOI] [PubMed] [Google Scholar]
  7. Del‐Carpio Munoz, F. , Powell, B. D. , Cha, Y. M. , Wiste, H. J. , Redfield, M. M. , Friedman, P. A. , & Asirvatham, S. J. (2013). Delayed intrinsicoid deflection onset in surface ECG lateral leads predicts left ventricular reverse remodeling after cardiac resynchronization therapy. Heart Rhythm: The Official Journal of the Heart Rhythm Society, 10(7), 979–987. 10.1016/j.hrthm.2013.03.045 [DOI] [PubMed] [Google Scholar]
  8. van Deursen, C. J. , Vernooy, K. , Dudink, E. , Bergfeldt, L. , Crijns, H. J. , Prinzen, F. W. , & Wecke, L. (2015). Vectorcardiographic QRS area as a novel predictor of response to cardiac resynchronization therapy. Journal of Electrocardiology, 48(1), 45–52. 10.1016/j.jelectrocard.2014.10.003 [DOI] [PubMed] [Google Scholar]
  9. Duckett, S. G. , Camara, O. , Ginks, M. R. , Bostock, J. , Chinchapatnam, P. , Sermesant, M. , … Rinaldi, C. A. (2012). Relationship between endocardial activation sequences defined by high‐density mapping to early septal contraction (septal flash) in patients with left bundle branch block undergoing cardiac resynchronization therapy. Europace, 14(1), 99–106. 10.1093/europace/eur235 [DOI] [PubMed] [Google Scholar]
  10. Gjesdal, O. , Remme, E. W. , Opdahl, A. , Skulstad, H. , Russell, K. , Kongsgaard, E. , … Smiseth, O. A. (2011). Mechanisms of abnormal systolic motion of the interventricular septum during left bundle‐branch block. Circulation: Cardiovascular Imaging, 4(3), 264–273. [DOI] [PubMed] [Google Scholar]
  11. Lang, R. M. , Badano, L. P. , Mor‐Avi, V. , Afilalo, J. , Armstrong, A. , Ernande, L. , … Voigt, J. U. (2015). Recommendations for cardiac chamber quantification by echocardiography in adults: An update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. European Heart Journal of Cardiovascular Imaging, 16(3), 233–270. 10.1093/ehjci/jev014 [DOI] [PubMed] [Google Scholar]
  12. Lumens, J. , Ploux, S. , Strik, M. , Gorcsan, 3rd J. , Cochet, H. , Derval, N. , … Bordachar, P. (2013). Comparative electromechanical and hemodynamic effects of left ventricular and biventricular pacing in dyssynchronous heart failure: Electrical resynchronization versus left‐right ventricular interaction. Journal of the American College of Cardiology, 62(25), 2395–2403. 10.1016/j.jacc.2013.08.715 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Maass, A. H. , Vernooy, K. , Wijers, S. C. , van‘t Sant, J. , Cramer, M. J. , Meine, M. , … Van Gelder, I. C. (2017). Refining success of cardiac resynchronization therapy using a simple score predicting the amount of reverse ventricular remodelling: Results from the Markers and Response to CRT (MARC) study. Europace, 10.1093/europace/euw445. [DOI] [PubMed] [Google Scholar]
  14. Mafi Rad, M. , Wijntjens, G. W. , Engels, E. B. , Blaauw, Y. , Luermans, J. G. , Pison, L. , … Vernooy, K. (2016). Vectorcardiographic QRS area identifies delayed left ventricular lateral wall activation determined by electroanatomic mapping in candidates for cardiac resynchronization therapy. Heart Rhythm: The Official Journal of the Heart Rhythm Society, 13(1), 217–225. 10.1016/j.hrthm.2015.07.033 [DOI] [PubMed] [Google Scholar]
  15. Ploux, S. , Lumens, J. , Whinnett, Z. , Montaudon, M. , Strom, M. , Ramanathan, C. , … Bordachar, P. (2013). Noninvasive electrocardiographic mapping to improve patient selection for cardiac resynchronization therapy: Beyond QRS duration and left bundle branch block morphology. Journal of the American College of Cardiology, 61(24), 2435–2443. 10.1016/j.jacc.2013.01.093 [DOI] [PubMed] [Google Scholar]
  16. Ponikowski, P. , Voors, A. A. , Anker, S. D. , Bueno, H. , Cleland, J. G. , Coats, A. J. , … van der Meer, P. (2016). 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. European Journal of Heart Failure, 18:(8):891–8975. 10.1002/ejhf.592 [DOI] [PubMed] [Google Scholar]
  17. Sipahi, I. , Chou, J. C. , Hyden, M. , Rowland, D. Y. , Simon, D. I. , … Fang, J. C. (2012). Effect of QRS morphology on clinical event reduction with cardiac resynchronization therapy: Meta‐analysis of randomized controlled trials. American Heart Journal, 163(2), 260–267. e3. 10.1016/j.ahj.2011.11.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Stankovic, I. , Prinz, C. , Ciarka, A. , Daraban, A. M. , Kotrc, M. , Aarones, M. , … Voigt, J. U. (2016). Relationship of visually assessed apical rocking and septal flash to response and long‐term survival following cardiac resynchronization therapy (PREDICT‐CRT). European Heart Journal of Cardiovascular Imaging, 17(3), 262–269. 10.1093/ehjci/jev288 [DOI] [PubMed] [Google Scholar]
  19. Stankovic, I. , Prinz, C. , Ciarka, A. , Daraban, A. M. , Mo, Y. , Aarones, M. , … Voigt, J. U. (2017). Long‐Term Outcome After CRT in the Presence of Mechanical Dyssynchrony Seen With Chronic RV Pacing or Intrinsic LBBB. JACC: Cardiovascular Imaging, 10(10 Pt A), 1091–1099. [DOI] [PubMed] [Google Scholar]
  20. Strauss, D. G. , & Selvester, R. H. (2009). The QRS complex–a biomarker that “images” the heart: QRS scores to quantify myocardial scar in the presence of normal and abnormal ventricular conduction. Journal of Electrocardiology, 42(1), 85–96. 10.1016/j.jelectrocard.2008.07.011 [DOI] [PubMed] [Google Scholar]
  21. Strauss, D. G. , Selvester, R. H. , & Wagner, G. S. (2011). Defining left bundle branch block in the era of cardiac resynchronization therapy. American Journal of Cardiology, 107(6), 927–934. 10.1016/j.amjcard.2010.11.010 [DOI] [PubMed] [Google Scholar]
  22. Surawicz, B. , Childers, R. , Deal, B. J. , Gettes, L. S. , Bailey, J. J. , Gorgels, A. , … Wellens, H. (2009). AHA/ACCF/HRS recommendations for the standardization and interpretation of the electrocardiogram: Part III: Intraventricular conduction disturbances: A scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Endorsed by the International Society for Computerized Electrocardiology. Journal of the American College of Cardiology, 53(11), 976–981. 10.1016/j.jacc.2008.12.013 [DOI] [PubMed] [Google Scholar]
  23. Sweeney, M. O. , van Bommel, R. J. , Schalij, M. J. , Borleffs, C. J. , Hellkamp, A. S. , & Bax, J. J. (2010). Analysis of ventricular activation using surface electrocardiography to predict left ventricular reverse volumetric remodeling during cardiac resynchronization therapy. Circulation, 121(5), 626–634. 10.1161/CIRCULATIONAHA.109.894774 [DOI] [PubMed] [Google Scholar]
  24. Yancy, C. W. , Jessup, M. , Bozkurt, B. , Butler, J. , Casey, Jr D. E. , Drazner, M. H. , … Wilkoff, B. L. (2013). 2013 ACCF/AHA guideline for the management of heart failure: Executive summary: A report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation, 128(16), 1810–1852. 10.1161/CIR.0b013e31829e8807 [DOI] [PubMed] [Google Scholar]

Articles from Annals of Noninvasive Electrocardiology : The Official Journal of the International Society for Holter and Noninvasive Electrocardiology, Inc are provided here courtesy of International Society for Holter and Noninvasive Electrocardiology, Inc. and Wiley Periodicals, Inc.

RESOURCES