Myocardial work in children with Wolff-Parkinson-White syndrome

Wolff-Parkinson-White Syndrome (WPW) has been associated with reduced local myocardial deformation, and when left ventricular dysfunction is present, catheter ablation of the accessory pathway may be required, even in asymptomatic patients. We aimed to evaluate the diagnostic value of non-invasive myocardial work in predicting subtle abnormalities in myocardial performance in children with WPW.Seventy-five paediatric patients (age 8.7 ± 3.5 years) were retrospectively recruited for the study: 25 cases with manifest WPW and 50 age- and sex- matched controls (CTR). Global myocardial work index (MWI) was measured as the area of the left ventricle (LV) pressure-strain loops. From MWI, global Myocardial Constructive Work (MCW), Wasted Work (MWW), and Work Efficiency (MWE) were estimated. In addition, standard echocardiographic parameters of LV function were evaluated. Despite normal LV ejection fraction (EF) and global longitudinal strain (GLS), children with WPW had worse MWI, MCW, MWW, and MWE. At multivariate analysis, MWI and MCW were associated with GLS and systolic blood pressure, and QRS was the best independent predictor of low MWE and MWW. In particular, a QRS > 110 ms showed good sensitivity and specificity for worse MWE and MWW values. In children with WPW, myocardial work indices were found significantly reduced, even in the presence of normal LV EF and GLS. This study supports the systematic use of myocardial work during the follow-up of paediatric patients with WPW. Myocardial work analysis may represent a sensitive measure of LV performance and aid in decision-making.


Introduction
In Wolff-Parkinson-White syndrome (WPW), the ventricles are electrically and mechanically pre-excited through an abnormal tongue of conductive tissue, the Kent accessory pathway (AP), connecting atria and ventricles, resulting in an asynchronous spread of ventricular depolarization [1]. Therefore, an asynchronous contraction with systolic and diastolic dysfunction of the left ventricle (LV) may result. Asymptomatic patients are supposed to have a benign course [2], however, when dysfunction of the LV is established, catheter ablation of the AP may be required [3].
Standard echocardiographic parameters and speckle tracking echocardiographic indices of LV function are load-dependent and may be inaccurate in specific settings [4]. Thus, myocardial work has been proposed as a new measure of LV myocardial performance, taking into account LV deformation and LV afterload [5], and giving an estimate of myocardial energetics [6]. Myocardial work has been previously used in children to calculate the area within the pressure-volume loop during one cardiac cycle [7][8][9]. Even though it has been invasively estimated in the past decades, a non-invasive method has been recently proposed. Combining longitudinal strain with standardized LV pressure curves, global and segmental myocardial 1 3 work can be evaluated using brachial cuff pressure and valvular time events [6].
We hypothesized that children with WPW syndrome, with normal standard echocardiographic parameters of LV function, may have an impairment of myocardial work indices, and thus, we aimed to use these indices to evaluate the presence of a subtle alteration in LV myocardial performance.

Study population
This is an observational study on consecutive Caucasian pediatric patients (age < 16 years old) who were diagnosed with WPW and followed up between December 2020 and May 2022 in the outpatients Pediatric Arrhythmic Clinic at Azienda Ospedaliera dei Colli -Monaldi Hospital, Naples, Italy. The study was carried out according to the STROBE checklist [10].
We retrospectively collected patients' data between September 2021 and May 2022. Only patients with a manifest WPW pattern on the surface electrocardiogram (ECG) were considered eligible for this study. Patients with combined structural heart diseases were excluded from the analysis. The control group was represented by healthy children who were referred for chest pain or palpitation and revealed to have normal cardiac anatomy and function, with no sign of WPW pattern on surface ECG. The matching of patients with controls was done individually, based on sex and age. Each patient with WPW was matched with two control children.
Demographic and clinical data were obtained from the patient's medical records. Informed consent was obtained from the patient's parents or legal guardians before the enrolment when data were collected retrospectively. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the local Ethics Committee.

Electrocardiogram
All patients underwent a complete 12-lead surface ECG at the time of echocardiography examination. The main electrocardiographic intervals were evaluated in both WPW and control groups. The PR and QRS intervals for each ECG were determined by the largest duration in any lead. The ECG algorithm by Arruda et al. [11] was used to discriminate septal, right free-wall, and left free-wall AP.

Echocardiographic examination
All patients underwent a complete transthoracic echocardiographic examination according to a standard protocol [12], using a GE E95 or a GE E80 ultrasound system (GE Healthcare, Wauwatosa, WI) with electrocardiographic tracing. Echocardiograms with inadequate images of LV myocardium were excluded from the analysis.
From the apical window, we recorded two-dimensional (2D) 4-chamber, 3-chamber, and 2-chamber views with a frame rate ≥ 60 frames/sec and optimized grayscale, then, the uncompressed images were transferred to a dedicated workstation (EchoPAC version 204, GE Healthcare) and analysed offline by one physician blinded to other clinical information (N.B.). To calculate the LV global longitudinal strain (GLS), the endocardium contour was semiautomatically traced and then manually adjusted as needed in each of the three apical views. The region of interest, between the endocardial and epicardial borders, was recognized by the EchoPAC software and, then, manually adjusted to ensure that the wall thickness was incorporated in the analysis, avoiding the pericardium. Results of segmental and global LV longitudinal peak systolic strain values were thus provided by the software. A 17-segment model was used for the study purpose. For all the enrolled patients, the image quality was excellent, and no LV segments were excluded from the analysis. GLS was reported as an absolute value throughout the study.
From 2D 4-chamber and 2-chamber apical views, we also evaluated LV Ejection Fraction (EF) by Simpson's method. Once the endocardial border was traced in enddiastole and end-systole, the echo machine software automatically calculated the EF value.
As previously described and validated by Russell et al. [6], myocardial work was estimated as a function of time throughout the cardiac cycle by combining non-invasively estimated LV pressure curve with LV strain measurements.
Myocardial Work Index (MWI) and derived indices [Myocardial Constructive work (MCW), Myocardial Wasted work (MWW), and Myocardial work efficiency (MWE)] were estimated using dedicated software (EchoPAC, version 204, GE-Healthcare). During echocardiographic acquisition, blood pressure was simultaneously measured by a cuff manometer.
The peak LV systolic pressure was assumed to be equal to the peak blood pressure measured by the cuff manometer and assumed to be uniform throughout the ventricle. The non-invasive LV pressure curve was then acquired using an empiric and normalized reference curve that was adapted to the length of the isovolumetric and ejection phases, defined by the timing of aortic and mitral valve events by echocardiography [6,[13][14][15].
Using the onset of the R wave on the ECG, strain and pressure measurements were coordinated. Myocardial work was then quantified by calculating the rate of segmental shortening by differentiation of the strain curve and multiplying this value with instantaneous LV pressure. This product is a measure of instantaneous power, which was integrated over time to obtain myocardial work as a function of time in systole, which is defined as the time interval from mitral valve closure to mitral valve opening.
At the time of the LV ejection phase, the work done by the myocardium during segmental elongation produces a loss of energy, which represents the MWW of that segment. Myocardial work obtained during segmental shortening represented the MCW of that segment. At the time of LV isovolumetric relaxation, there is an inversion of those definitions so that myocardial work during shortening represents segmental MWW, and work during lengthening represents segmental MCW. By averaging segmental MCW and MWW, global MCW and MWW were estimated for the entire LV. MWE represented the constructive work divided by the sum of constructive and wasted work (0-100%) [6,[13][14][15].

Statistical analysis
Continuous variables were tested for normality using the Shapiro-Wilk method, and expressed as median and interquartile ranges (IQR) or mean and standard deviation (SD) according to their distribution. Categorical variables were presented as frequency (%).
Comparisons were obtained by unpaired T-test and Mann-Whitney U test. The Chi-square test was used to compare categorical variables.
Multivariate linear regression models were generated to identify factors that could independently justify worse values of Myocardial Work (dependent variable). Relevant parameters (QRS interval, age, sex, EF, GLS, sBP), as well as the diagnosis of WPW, were included in all models. The normality of residuals was assessed by the Shapiro-Wilk test.
In patients with WPW, the receiver operating characteristic (ROC) curve and area under the curve (AUCs) were used to identify a threshold value of QRS associated with worse Myocardial Work indices.
We measured also MWI and GLS intra-and interobserver variability by interclass correlation (ICC) and coefficient of variation (CV) in 10 randomly selected echocardiographic exams (5 exams from the WPW group and 5 exams from the control group). To evaluate intra-observer variability, the same observer (N.B.) who performs all the echocardiographic measurements reassessed the same parameters at least four weeks later to avoid recall bias. To evaluate inter-observer variability, another observer (G.D.C.), independent and blinded to the original measurements, evaluated the same 10 echocardiographic exams.
Statistical significance was defined as a two-tailed p-value < 0.05. Statistical analysis was performed using MedCalc, version 18.11 (MedCalc Software, Mariakerke, Belgium) and R version 4.0.5 (R Project for Statistical Computing, Vienna, Austria).

Study population
Within the study period, a total of 40 patients (65% males, age 8.9 ± 3.8 years) with manifest WPW at surface ECG were referred to our Institution. The echocardiographic studies of 15 patients (67% males, age 9.7 ± 5.0 years) presented inadequate myocardium visualization or absent recording of the required echocardiographic views for speckle-tracking analysis and were excluded from the study. One male patient was excluded for combined Ebstein anomaly. After control matching, a final cohort of 75 patients (25 patients with WPW and 50 controls) was included in the study. The cohort included 48 male patients (64%). The majority of children with WPW presented a septal AP (19 patients), 4 patients presented a right free-wall AP, and two patients presented a left free-wall AP. Echocardiograms were obtained at a mean age of 8.7 ± 3.5 years. Patient demographic and standard echocardiographic characteristics are summarized in Table 1. Six patients in the WPW group were on antiarrhythmic drugs (flecainide in 5 patients and flecainide plus sotalol in one patient) with persistent pre-excitation signs on surface ECG. There were no differences in QRS or Myocardial Work indices between patients on anti-arrhythmics and naïve patients. No significant differences were found between the WPW group and the control group in terms of age, sex, height, weight, body surface area, heart rate, and sBP. Diastolic blood pressure was lower in the WPW group compared to the control group. LV EF and GLS were within the normal range in both groups and no significant differences were found between groups (Table 1). At the surface ECG, all patients in the control group presented a QRS interval within the normal limit for the age (70-80 ms), while all patients in the WPW group presented a wide QRS for the age [110 (100-120) msec].

Myocardial work analysis
Myocardial Work analysis is shown in Tables 2 and 3. Global MWI, MCW, and MWE were significantly reduced in the WPW group compared to the control group (Fig. 1). Accordingly, MWW was significantly increased in the WPW group compared to the control group.

Multiple linear regression model
Multivariate regression models were applied to the entire studied population to further investigate parameters that may predict worse myocardial work indices (Table 4). Low MWI and MCW were associated with low GLS and low sBP. Low MWI was also associated with younger age. A wider QRS was the only predictor of higher values of MWW and lower values of MWE.
In the WPW group, the AUC of the QRS interval was analysed, and a cut-off > 110 ms showed a 100% sensitivity and 74% specificity for lower IQR of MWE (AUC of 0.84) and a 100% sensitivity and 70% specificity for higher IQR of MWW (AUC of 0.82) (Fig. 2).

Reproducibility analysis
The inter-observers agreement was very good for MWI (ICC 0.80, CV 5.5%) and GLS (ICC 0.79, CV 2.6%). Intraobserver reproducibility was also good for MWI, with an ICC of 0.78 and CV of 4.3%, and moderately good for GLS, with an ICC of 0.70 and of CV 3.6%.

Discussion
For the first time, the present study assessed myocardial performance in a pool of children with WPW by using noninvasive Myocardial Work analysis. Our findings suggested that children with WPW may have subtle impairment of  [16][17][18] and adult [19,20] patients with WPW with different imaging modalities, even in the absence of tachycardia. Recently, Nagai et al. [20] used speckle-tracking echocardiography to demonstrate an impairment of LV GLS, global circumferential strain, and LV EF in a population of adult patients with WPW (mean age 40 ± 12 years). In a younger population (mean age 14.1 years -range 6.9-21.6 years), Akimoto et al. [21] demonstrated, instead, a reduction in mid-basal circumferential strain but not in LV EF. The same authors [21] also demonstrated low GLS in WPW patients with right-sided AP, however, patients with left-sided AP presented normal GLS. In agreement with Akimoto et al., in our study, children with WPW had normal values of LV EF compared with controls. Similarly, in our very young cohort of WPW patients (mean age 8.7 ± 3.5 years), GLS did not differ significantly from controls. As previously hypothesized [18], indeed myocardial dysfunction may appear with longer exposure to AP and be more frequently found with advancing age. However, we found an impairment of all myocardial work indices in WPW patients compared to controls. We hypothesized, in the presence of normal LV EF and GLS, myocardial work indices may represent a more sensitive method to study myocardial performance in paediatric patients with WPW. GLS indeed may detect subtle myocardial impairment even in the presence of normal LV EF [22], but it still carries load-dependency limitation [23]. Myocardial work has been shown to have an added value, taking into account deformation as well as afterload, and being less affected by loaddependence [24,25].
In accordance with previous studies, we found low MWI and MCW associated with low GLS and sBP 8 ; moreover, low MWI was found associated with younger age [26,27].
In a study by Maréchaux [28], in WPW patients with interventricular septum accessory pathway, as well as in patients with left bundle branch block (LBBB), a specific LV longitudinal strain pattern was described, which, interestingly, was more frequently found in patients with prolonged QRS duration. In normal conditions, myocardial segments contract all together during systole, with very little or no waste of energy. It is observed that the presence of myocardial dyssynchrony reduces myocardial performance because work performed by one segment is wasted by stretching other segments [29].
In line with these observations, we found a correlation between impaired MWE and MWW and prolonged QRS duration.
We speculated these observations may be related to the higher dyssynchrony carried by a prolonged QRS observed in children with WPW. An impairment of MWE and MWW has been previously demonstrated in patients with cardiomyopathy and LBBB [29,30]. Similarly, in our patients with WPW, the presence of pre-excited myocardial segments shortening against a closed aortic valve may be responsible for the observed reduced work. As a consequence, the work done by the pre-excited segments does not contribute to LV EF but may represent a measure of contractile reserve after AP ablation, similar to patients with LBBB who undergo cardiac resynchronization therapy [29].
Interestingly, in our study, the impairment of myocardial work was not independently related to a diagnosis of WPW per se. It means that not all patients with WPW would benefit from AP ablation, but rather only patients with wider QRS and worse values of myocardial work. In particular, we found children with WPW and QRS > 110 ms at higher risk for impairment of myocardial performance (worse IQR of MWE and MWW).
The location of AP is generally associated with different QRS duration, with the left-sided AP having a shorter QRS than the right-sided AP [31]. Although we did not analyse myocardial work based on AP location considering the relatively small WPW group, it is notable that the two patients with left free-wall AP presented high value of MWI, MCW, MWE, and low value of MWW.
As demonstrated previously, evaluation of segmental work in ventricles with asynchronous contraction may contribute to the understanding mechanism underlying remodelling and developing LV dysfunction [29,32]. In accordance, myocardial work analysis should be regularly performed during follow-up of patients with WPW, and in the presence of QRS interval > 110 ms, a closer follow-up should be considered to detect asymptomatic children with reduced values of myocardial work indices who may benefit from an early AP ablation.

Limitations
This study has some limitations. The lack of outcome information after the electrophysiological study represented the main one. As a result, we did not have data on myocardial work trends after AP catheter ablation. Moreover, considering the relatively small population of WPW patients, we did not perform a subgroup analysis to investigate peculiarities between different AP locations. Although the left-sided atrio-ventricular connection is the most frequent AP location, a greater proportion of the analysed children had septal AP. Indeed left-sided AP more frequently presents a non-manifest or concealed anterograde conduction, with no visible or minimally visible ECG signs of ventricular preexcitation. Considering the majority of patients were asymptomatic, and referred to our out-patients clinic following a routine visit, this higher rate of septal AP was attributed to the higher occurrence of manifest pre-excitation ECG signs and usual broader delta-wave associated with this site.
Finally, considering the study was conducted on children, we did not have data on the use of myocardial work on adults with longer exposure to the AP.

Conclusion
In pediatric patients with WPW, we found myocardial work to detect subtle impairment of myocardial performance even in the presence of normal LV EF and GLS. In this study, lower MWE and higher MWW were related to a prolonged QRS, with QRS interval > 110 ms having a higher likelihood of reduced myocardial work.
Our study supports the systematic use of myocardial work during the follow-up of patients with WPW, considering the insight it can provide in understanding LV remodelling and dysfunction. If confirmed in larger studies, our findings may have value in identifying a subset of paediatric patients with WPW who may benefit from a closer follow-up and, eventually, more proactive electrophysiological management.