Relation of atrial electromechanical delay to P‐wave dispersion on surface ECG using vector velocity imaging in patients with hypertrophic cardiomyopathy

Abstract Objectives Heterogeneity of structural and electrophysiologic properties of atrial myocardium is common characteristic in hypertrophic cardiomyopathy (HCM). We assessed the dispersion of atrial refractoriness on surface ECG using P‐wave dispersion (PWD) and its relation to atrial electromechanical functions using vector velocity imaging (VVI) in HCM population. Methods Seventy‐nine HCM patients (mean age: 43.7 ± 13 years, 67% male) were compared with 25 healthy individuals as control. P‐wave durations, Pmax and Pmin, P‐wave dispersion (PWD), and P terminal force (PTF) were measured from 12‐lead ECG. LA segmental delay (TTP‐d) and dispersion (TTP‐SD) of electromechanical activation were derived from atrial strain rate curves. Results HCM patients had longer PR interval, PW duration, higher PWD, PTF, QTc compared to control (p < .001). HCM patients were classified according to presence of PWD into two groups, group I with PWD > 46 ms (n = 25) and group II PWD ≤ 46 ms (n = 54). Group I showed higher prevalence of female gender, higher PTF, QTc interval, left ventricular outflow tract (LVOT) obstruction, p < .01, LVOT gradient (p < .001), LV mass index (p < .01), E/E' (p < .01), and severe mitral regurgitation (p < .001). Moreover, PWD was associated with increased atrial electromechanical delay (TTP‐d) and LA mechanical dyssynchrony (TTP‐SD), p < .001. LA segmental delay and dispersion of electromechanical activation were distinctly higher among HCM patient. Conclusion PWD is simple ECG criterion, and it is associated with more severe HCM phenotype and LA electromechanical delay while PTF is linked only to atrial remodeling.


| INTRODUC TI ON
Hypertrophic cardiomyopathy (HCM) is a genetic heart disease with heterogeneous clinical expression, outcome, and management options (Maron et al., 2014;Maron et al., 2016). Left atrial (LA) remolding and dysfunction, (Maron et al., 2016) atrial fibrosis (Maron et al., 2014), and LA appendage dysfunction (Maron et al., 2015) have been described. They are recognized as common, important consequences (Maron et al., 2015) and known to lead to electro-anatomical remodeling, alter the atrial conduction properties and act as a substrate for the occurrence of AF (Rowin et al., 2017). P-wave dispersion (PWD), defined as the difference between the maximum and minimum P-wave duration on surface ECG, is a new electrocardiographic marker that has been associated with inhomogeneous and discontinuous propagation of sinus impulses (Magnani et al., 2010). The correlation between the presence of interatrial and intra-atrial conduction abnormalities and the induction of paroxysmal atrial fibrillation (AF) has been well documented (Liu et al., 1998). The estimation of the probability of patients in developing paroxysmal AF might guide the clinician in the management and stratification of HCM patients at higher risk of developing AF.
Recently, strain and strain rate measured by novel vector velocity imaging (VVI) have been used in evaluating cardiac mechanics. It allows simultaneous and precise analysis of atrial mechanics during its different phases reservoir, conduit, and contractile functions (Badran et al., 2019;Pala et al., 2010;Sanders et al., 2003) in addition to measurement of atrial electromechanical delay (Acar et al., 2009;Rein et al., 2003).
Our hypothesis was that there might be a relationship between LA mechanical changes in HCM and the atrial electrophysiologic characteristics on the 12 lead ECG. In the present study, we investigated the relation of PWD on surface ECG to LA electromechanical functions using VVI in patients with HCM.

| Study population
Between January 2018 and May 2019, 79 consecutive HCM patients were referred to our echocardiographic laboratory for different purposes (diagnosis, risk stratification, regular follow-up) and included in the study. They were examined in a single center (Yacoub Research Unit, Menoufia University, Egypt, and as a part of the BA-HCM National Registry Program). This is a cross-sectional study in which patients with established diagnosis of HCM were included after their informed consent and after approval of ethical committee of Menoufia University. The diagnosis of HCM was based on echocardiographic demonstration of a nondilated, hypertrophic left ventricle (LV) with increased LV wall thickness (≥15) in one or more LV segments and in the absence of any identifiable causes capable of producing such hypertrophy (Maron et al., 2014(Maron et al., , 2015(Maron et al., , 2016. All patients were in sinus rhythm.

| Exclusion criteria
Patients with AF, history of paroxysmal atrial fibrillation, coronary artery diseases, prior surgical myectomy, prior pacemaker, or implantable cardiovertor defibrillator (ICD) implantation, and patients with known phenocopies of HCM (e.g., Fabry disease, lysosomal associated membrane protein-2 cardiomyopathy, or amyloidosis) or poor quality echocardiographic imaging were excluded from the study.

| Electrocardiographic analysis
Twelve leads surface ECG was recorded for all patients at the study entry with a paper speed of 25 mm/s and signal size of 10 mm/mv. ECGs were scanned using a high-resolution scanner, transferred into F I G U R E 1 Measurement of P-wave duration from computer screen using caliper computer memory, magnified and P-wave duration was manually measured from the commencement to the end of the P wave, using electronic caliper, on a high-resolution computer screen. P-wave start and end points were defined at the junction between P-wave deflection and the isoelectric line. To achieve greater precision in measuring P-wave dispersion, we measured simultaneous digital recording of all 12 ECG leads and on computer screen with the high zoom capabilities, Figure 1. P-wave dispersion is defined as the difference between maximum and minimum P-wave durations measured at all ECG derived leads (Acar et al., 2009;Badran et al., 2019;Liu et al., 1998;Magnani et al., 2010;Pala et al., 2010;Rein et al., 2003;Sanders et al., 2003;Tosun et al., 2018). Intraobserver and interobserver coefficients of variation were found to be 4.1% and 4.4% for PWD, respectively (Badran et al., 2019;Pala et al., 2010;Tosun et al., 2018

| Conventional echocardiography
Echocardiographic examination was performed in the left lateral decubitus in the apical 2 & 4 chamber, parasternal long, short axis views using standard transducer positions. Esaote Mylab Gold 30 ultrasound system (Esaote S.p.A, Florence, Italy) equipped with a multi-frequency 2.5-3.5 MHz phased-array transducer was utilized.
Color flow and continuous-wave Doppler were used to define resting LV outflow tract gradient (LVOT) and to estimate pulmonary artery pressure (PAP) from tricuspid regurgitation velocity (Bernolli equation). Peak early (E) and late (A) trans-mitral filling velocities were measured from Doppler mitral inflow. Longitudinal peak systolic (S'), early diastolic (E'), and atrial diastolic (A') velocities as well as isovolumetric relaxation time (IRT) were obtained by placing a pulsed wave tissue Doppler (TDI) sample volume at the lateral mitral annulus in the apical four-chamber view. E/E' was derived from the ratio between E of mitral inflow velocity and E' of annular velocity using TDI.

| Analysis of LA deformation
Border tracking of the LA was manually traced from the digitized 2D video clips recorded during breath holding and with good quality ECG signals, which were acquired and stored for off-line analysis using X-Strain software with a frame rate between 40 and 80 fps.
The "Zoom/RES" feature on the echocardiographic machine was used to improve the accuracy of atrial measurements. A circular region of interest was traced on the LA endocardial cavity interface in apical four-chamber view at end diastole (LA minimum cavity area) using a point-and-click approach. Time-volume curves were extracted from LA wall tracking that provided automatically indexed maximum and minimum LA volume and left atrium ejection force (LAEF) (Cameli et al., 2016).
Definition of the LA endocardial border enabled the system to calculate regional longitudinal deformation of the LA walls. Peak systolic strain (ε sys ) and LA systolic SR (SR sys ) were measured as a positive curve at LV systole (representing reservoir function), early diastole (SR e ) (representing conduit function), and late diastole (SR a ) (representing contractile function). Image processing algorithm automatically subdivides the atrial wall into 12 segments distributed in septum and lateral and posterior LA wall-"roof." The graphs for each segment were displayed and averaged to calculate global LA

functions.
To estimate LA mechanical dyssynchrony, the index of myocardial systolic activation was calculated from regional strain rate curves for each segment, as time from the beginning of P wave of ECG to the peak of SR a wave (TTP). LA electromechanical delay was measured as the difference between TTP (d-TTP) in 12 LA segments (difference between the longest and shortest cycle). LA mechanical dyssynchrony was defined as the standard deviation of the time-topeak SR a (TTP-SD).

| Inter-and Intraobserver variability
Two independent observers performed two separate quantitative εsys and SR analyses of LA images blindly in 35 participants.

| Statistical analyses
Data were presented as numbers (%) or as mean and standard deviation values. The distribution of qualitative variables was analyzed by chi-square test or Fisher's exact test. Quantitative variables were correlated by the use of Pearson's correlation coefficient "r". All tests were two-tailed, and p-value <.05 was considered statistically significant. Receiver operating characteristic (ROC) curve analysis was performed to select optimal cutoff values of LA deformation measurements. The analysis was performed by the IBM SPSS statistics software package.

| Clinical characteristics of the study population
Demographic and clinical characteristics of the study population were depicted in Table 1. There was no significant difference between patients and control subjects with respect to age, gender, BSA, HR, SBP, or DBP. HCM patients were predominantly nonobstructive 61 (77.2%) with LVOT PG < 30 mmHg, 59 (75%) had asymmetric left ventricular hypertrophy (LVH), 61 (77.2%) patient were symptomatic, and 23 (29%) were NYHA class III and IV.

| Changes in surface ECG in the study population
Almost all electrocardiographic variables showed significant difference in comparison with healthy individuals. P-wave duration, PWD, and P terminal force were significantly higher in HCM patient (92.4 ± 14.2, 42.5 ± 15.4, 66.2 ± 48.9) versus controls (65.5 ± 9.6, 26.2 ± 8.9, 25.2 ± 8.76) p < .0001, <.008, <.0001, respectively. PTF was verified in 58 (73.4%) of HCM patients versus 16 (64%), in control group p < .01. Also PR interval was significantly prolonged in HCM patients compared with the control group (164.3 ± 71.3 vs. 133.7 ± 14.9 ms), p < .001. QT intervals & QT dispersion (p < .0001) and QRS duration (p < .01) were significantly prolonged in comparison with control group. Voltage criteria for LVH and repolarization abnormalities were significantly prevalent in HCM patients (p < .001). On the contrary, the prevalence of LBBB in HCM did not reach significant difference when compared with healthy individuals.

TA B L E 1 Clinical and electrocardiographic characteristics
Furthermore, comparing LA mechanics between HCM patients with prolonged PWD and those with normal PW disper- 39.9 ± 32.3, p < .01) compared with patients with normal PW dispersion ( Figure 4).

| Correlates of ECG variables in patient population (table 4)
Taking the HCM patients altogether, Pearson's correlation coefficient was utilized to study the interaction between ECG variables functional status and LA structure and function. No significant correlation was noted between NYHA class and PWD (r = .11, p < .36), PTF (r = .14, p < .32), or PR interval (r = .13, p < .25), respectively. P terminal force was inversely correlated with LA volume and left atrial volume index (Figure 3) (Gersh et al., 2011;Maron et al., 2002).

| D ISCUSS I ON
Occurrence of AF initiate further clinical deterioration, thromboembolic complications, and increase mortality in this population . Therefore, early identification of HCM patients who are at a higher risk to develop AF is critical to prevent its harmful consequences.
Numerous studies have focused on predictors of AF in patients with HCM. Advancing age, functional status, LA enlargement, P-wave duration, and extent of LV fibrosis have been suggested to predict the propensity to develop AF in these patients (Cecchi et al., 1997;Yamaji et al., 2001).
Several investigators studied the changes in P-wave related entities for instance, coronary artery diseases, coronary artery bypass graft, hypertension, heart failure, and paroxysmal AF (Dilaveris et al., 1999;Frost et al., 1996;Yamada et al., 2000). Recently, PWD has been proposed to be a useful index for identifying AF. Abbreviations: LAD, left atrium; LAV, left atrial volume; LAVI, left atrial volume index; ESD, end systolic dimension; EDD, end diastolic dimension; FS, fractional shortening; EF, ejection fraction, PAP, pulmonary artery pressure, MWT, maximal wall thickness, LVPW, left ventricular posterior wall, LVMI, left ventricular mass index, LVOT, left ventricular outflow tract; E:early mitral inflow velocity; A:atrial mitral inflow velocity, DT, deceleration time, Ea, mitral annulus early diastolic velocity. εsys, peak systolic strain, TTP, time-to-peak strain, TTP-SD, standard deviation of time-to-peak strain, SR sys , peak systolic strain rate, SR e , early diastolic strain rate, SR a , atrial diastolic strain rate. electrocardiographic fragmentation, and an increased dispersion of atrial refractoriness (Abhayaratna et al., 2008;Suga, 1974;Tuluce et al., 2014). These properties reflected in P-wave morphology and duration on the 12-lead ECG. Electromechanical inhomogeneities of atrial myocardium in HCM population are usually associated with generation of unidirectional block of premature impulses, hence trigger atrial reentry (Kose et al., 2003).
In our study, we tried to find the relation between ECG derived atrial electrical parameters like PWD, PTF, and atrial mechanical function using VVI that precisely measure atrial phasic functions and electromechanical delay.
In the present study, 32% of HCM patients had significantly increased dispersion of P-wave duration; both maximum and minimum P-wave durations were significantly different from those with no dispersion. Higher PWD on the surface ECG probably thought to reflect the heterogeneity of mechanical and electrophysiologic properties of atrial myocardium.
In our study, LA diameter and LAVI were correlated directly with PTF that reflects particularly atrial remodeling, while PWD was associated more with ventricular structure and the severity of HCM phenotype. Indeed, patients with PWD showed increased LVMI, higher LVOT gradient, more severe mitral regurgitation, and the  (Tuluce et al., 2015).
Furthermore, atrial electromechanical dyssynchrony was significantly manifest in HCM with PWD but was not correlated with PTF.
This can be verified by the fact that the extremely anisotropic properties of the atrial myocardium due to microarchitectural changes in HCM might be more crucial in the genesis of AF paroxysms than cavity size (Lavergne et al., 1986;Misier et al., 1992). Fananapazir et al. (1989) investigated 155 HCM patients using electrophysiologic study, and they reported that 66% had prolonged sinoatrial conduction time. The presence of prolonged and fractionated atrial electrograms revealed a strong association with paroxysmal AF induction.
The importance of prolonged PWD in distinguishing HCM patients who are prone to AF risk from those without was previously investigated; however, the value of PWD in predicting future development of AF in a follow-up study has not been determined yet. Kose et al. (2003) categorized patients with HCM into two groups: group I with and group II without paroxysmal AF attacks, and compared their PWD to identify the value of PWD in prediction of AF.
In their study, they demonstrated a value of 46 ms as a cutoff point of PWD that differentiates patients with previous PAF attacks with 76% sensitivity and 82% specificity.
Another study conducted to estimate predictors of AF in HCM included 27 patients with HCM with a previous history of documented AF attack and 53 patients who had no such history (Ozdemir et al., 2004). This study demonstrated a PWD value of >52.5 ms differentiate patients from controls with a sensitivity of 96%, a specificity of 91%, and a positive predictive accuracy of 84%.
In our study, we used comparable PWD cutoff value derived from 2SD of mean value of PWD in healthy individuals. Our study had a similar number of patients compared to their patient number; however, our study not designed as a follow-up study.
Moreover, our study HCM patients with PWD had no larger LA size but higher LV end diastolic pressure as estimated by E/E' and more electromechanical delay. Our study confirm the findings by Tuluce et al. (2015), and they illustrated that electrical LA remodeling and not structural impairment has the main role in promoting AF.
ECG would be a straightforward, inexpensive, readily accessible, and noninvasive means to diagnose atrial electromechanical dyssynchrony if reliable criteria are available. PWD can be used as bed side prognostic test that able to discriminate HCM patients with atrial electrical heterogeneity at higher risk to develop AF. The criterion used in the present study is one of the most frequently used clinically which is PWD. Increased PWD of left atrial electrical properties is thought to reflect the heterogeneity of atrial electromechanical properties and electrophysiologic parameter that reflect prolongation of intra-atrial and interatrial conduction time and the inhomogeneous propagation of sinus impulses. This confirms the findings of other studies (Flaker et al., 1995;Henry et al., 1976;Ozdemir et al., 2004;Turitto et al., 1998).

| Clinical implication
The phenotypic variability of HCM is not only limited to variability of severity and extent of myocardial hypertrophy, presence or absence of obstruction, presence of absence of mitral regurgitation but also rather includes a set of functional manifestations and the electrophysiologic properties Moon & McKenna, 2012;Soler et al., 2018). It is important to look for prognostic markers that identify patients more prone to adverse cardiovascular events. This study shows considerable importance of most of ECG parameters including P-wave duration, P-wave dispersion, PTF, and PR intervals, F I G U R E 6 Relation of PTF to left atrial volume index as part of electrical characteristics of the disease, and adds to clinical evaluation and risk stratification of HCM population.

| CON CLUS ION
We concluded that PWD is associated with more severe HCM phenotype and left atrial electromechanical delay, while PTF is linked to atrial remodeling. The ECG parameter, if properly ascertained, emerges to be a practical criterion of atrial electromechanical delay.
PWD is revealed as a valuable parameter of easy measurement that signifies a greater tendency to the development of supraventricular arrhythmias, particularly AF. P-wave related entities may be considered as ECG-based marker of LA remodeling and predictors of increased risk of AF occurrence in HCM patients. If echocardiography is not available, ECG can be helpful for estimation and quantification of LA size and electromechanical properties.

CO N FLI C T O F I NTE R E S T
All authors declare that there is No conflict of interest.