INTRODUCTION
Dilated cardiomyopathy (DCM), the most common form of pediatric cardiomyopathy, is a major cause of heart failure (HF) in children.1 It is characterized by left ventricular (LV) dilatation, LV systolic dysfunction, and intraventricular conduction delays or blocks. The annual incidence of DCM in children under 18 years old is reported to be 0.57 cases per 100 000.2 Left bundle branch block (LBBB), a characteristic pattern of ventricular electromechanical dyssynchrony, is recognized as an independent risk factor for cardiac mortality. Epidemiological studies have further established that LBBB is frequently associated with structural remodeling, LV dilatation, and LV ejection fraction (LVEF) reduction.3 Ekkehard et al.4 demonstrated that LBBB accounted for 30% portion in children with DCM, and was associated with cardiac death.
Cardiac resynchronization therapy (CRT) has been shown to improve outcomes in patients with HF and ventricular electromechanical dyssynchrony.5 In parallel, surgical pulmonary artery banding (PAB) has emerged as a potential intervention to improve cardiac function in infants with end‐stage DCM and preserved right ventricular (RV) function.6 Nevertheless, evidence regarding the combined use of CRT and PAB in infants with LBBB‐associated DCM remains scarce. Given the limited pediatric data, this study aimed to assess the short‐term outcomes and evaluate the treatment response to combined CRT and PAB in this unique patient subgroup.
METHODS
Ethical approval
This study was approved by the Ethics Committee of Beijing Fuwai Hospital (2022‐1859). Electronic informed consent forms were completed by the parents of all participants.
Patients
A retrospective study was conducted on five patients with LBBB‐associated DCM who underwent PAB and CRT implantation at Beijing Fuwai Hospital between March 2023 and August 2024 (Figures 1 and 2).
The enrollment criteria were as follows: 1) age ≤ 1 year old; 2) diagnosis of DCM (LVEF ≤30%) and LBBB; 3) normal RV function; 4) New York Heart Association (NYHA) functional class III–IV; 5) provision of informed consent and voluntary participation; and 6) availability of complete clinical data. Exclusion criteria included: 1) those with serious preoperative conditions such as multi‐organic insufficiency, severe infection, metabolic diseases and chromosomal genetic disorders; 2) secondary causes of DCM (hypertension, coronary artery disease, myocarditis, tachycardia‐induced cardiomyopathy, systemic or pericardial disease, and congenital heart disease); and 3) presence of cardiac fibrotic scar or interstitial fibrosis as identified on tissue characterization.
According to the predefined enrollment and exclusion criteria, five infants under 1 year of age with LBBB‐associated DCM were enrolled. All patients had an LVEF below 30% and normal RV function. Prior to surgery, all patients had received guideline‐directed medical therapy for over 2 months, with drug dosages carefully adjusted based on heart rate, blood pressure, urine output, and the severity of HF.
All five patients underwent combined PAB surgery and CRT device implantation, and continued to receive guideline‐directed medical therapy during the follow‐up period. Preoperative evaluation included electrocardiogram (ECG), echocardiography, cardiac magnetic resonance imaging (cMRI), computed tomography (CT), laboratory testing, and genetic analysis. After surgery, patients were regularly followed at 1, 3, and 6 months, with ECG, echocardiography, and N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) levels monitored and recorded at each visit.
Diagnostic criteria for LBBB
The diagnostic criteria were as follows: 1) Leads V5 and V6 demonstrate an “R” or “qR” pattern, with broad notched or slurred R wave; 2) Leads V1 and V2 show a “QS” or “rS” pattern; 3) QRS duration is ≥ 90 ms (under 1‐year old); 4) Presence of left axis deviation. (Routine 12‐lead ECG; speed, 25 mm/s; voltage, 10 mm/mV)5, 7
Diagnostic criteria for DCM
The diagnostic criteria included: 1) LV dilatation (LV internal dimension in diastole >112% corrected for age and body surface area, and the value above 117% increases the specificity). 2) LVEF reduction (fractional shortening [FS] <25% and/or LVEF <45%).8, 9
Evaluation of cardiac function
ECG, cMRI, and CT were performed in all patients. Echocardiographic measurement included the severity of mitral regurgitation (MR), pulmonary artery pressure, LV end‐diastolic dimension (LVEDd), LVEDd z‐score, LV end‐systolic dimension (LVESd), LVESd z‐score, LVEF, and RVEF.8, 9 cMRI provides functional assessment of myocardial performance (LV and RV). Tissue characterization of fibrotic scar, interstitial fibrosis, edema, and hyperemia was also evaluated.9 CT was used to delineate anatomic structures and identify coronary artery abnormalities.9
Evaluation of cardiac synchronization
Transthoracic echocardiograms were recorded before and after PAB and CRT. The pre‐ejection interval, anteroseptal‐post time to peak strain delay, septolateral time to peak strain delay, and longitudinal global strain were calculated at each follow‐up time point.
Guideline‐directed medical therapy
All children received regular oral digoxin (doses targeting lower serum digoxin concentrations was 0.5–0.9 ng/mL), β‐Receptor antagonists (heart rate: 90–110/min), diuretics [furosemide 1−2 mg/(kg·day)], spironolactone [1−2 mg/(kg·day)], angiotensin‐converting enzyme inhibitors [captopril: 0.5–0.9 mg/(kg·day)].10
Pulmonary artery banding
Open‐chest surgery was performed on the beating heart through a median sternotomy incision. The band, made by polytetrafluoroethylene, was carefully placed around the pulmonary arterial trunk and fixed with 6‐0 prolene suture, ensuring no compromise of systemic—and particularly coronary—blood flow. A systolic right ventricle pressure to systolic arterial pressure ratio of 0.6 was considered appropriate.6
Biventricular pacing‐CRT devices implantation
The biventricular pacing (BiVP)‐CRT devices were implanted through a median sternotomy. The right atrial lead was positioned in the right atrial appendage, the RV lead at the RV apex, and the LV lead on the surface of the LV free wall. The epicardial electrodes were tunneled into the subcutaneous abdomen.
The LV lead should be positioned to avoid phrenic nerve stimulation and obtain the narrowest biventricular paced QRS (Figure 2).11, 12
Study endpoints
The primary endpoint was improvement in clinical functional class and cardiac function, assessed by LVEF and serum levels of NT‐proBNP. The secondary endpoint was cardiac regeneration, evaluated through measurements of LVEDd, LVEDd z‐score, LVESd, LVESd z‐score, cardiac resynchronization, and QRS duration.
Statistical analysis
Continuous variables were summarized as median (range), and categorical variables were represented as frequency (percentage).
RESULTS
Baseline demographic characteristics
A total of five infants were enrolled in our study, including two females and three males. All patients had been diagnosed with LBBB‐associated DCM and HF for a median duration of 2 months (range: 2–10 months). The median age at surgery was 6 months (range: from 3–12 months), and the median weight at surgery was 7.0 kg (range: 4.8– 7.5 kg).
The median resting heart rate was 135 beats/min (range: 126 –152 beats/min). Genetic testing was negative in 3 patients; one patient carried a paternally inherited FLNC mutation, and another carried a maternally inherited TTN mutation. All patients were classified as NYHA functional class IV.
There was no tissue characterization of fibrotic scar or interstitial fibrosis on cMRI. Preoperative liver and kidney function, as well as pulmonary artery pressure, were within normal limits.
CRT parameters
The mean pacing parameters were as follows: threshold, 1.00 ± 0.31 V; sensing amplitude, 5.10 ± 1.11 mV; and impedance, 247 ± 36 Ω. All parameters remained stable during the 3‐ and 6‐month follow‐up visits.
PAB pressure gradient
After PAB placement, the ventricular septum shifted slightly leftward, and LVEDd decreased during surgery. The initial trans‐PAB pressure gradient was 20–25 mmHg. A substantial increase in this gradient was observed during the postoperative period, reaching a range of 40–60 mmHg by 3–6 months, with the highest pressure recorded at 6 months (Figure S1).
The primary endpoints
BiVP‐CRT combined with PAB resulted in a substantial improvement in clinical functional class and cardiac function. The functional class of five patients increased from NYHA IV to I. The median LVEF increased from 26% preoperatively to 65% at the last follow‐up.
At the 3‐month follow‐up, three patients (60%) had achieved normal cardiac function (LVEF ≥50%), and by 6 months, all five (100%) exhibited normal function (LVEF ≥60%). The NT‐proBNP level rose to a median of 6122 pg/mL (range: 5290–7824) on postoperative day 3, declined by 1 month to 3424 pg/mL (range: 1244–6061), and remained at a low level of 1860 pg/mL (range: 301–2111) at the 6‐month follow‐up (Table 1 and Figure S1).
The secondary endpoints
The median LVEDd, LVEDd z‐score, LVESd, and LVESd z‐score decreased at the 3‐ and 6‐month follow‐ups from baseline. Median QRS width decreased from 138 ms at baseline to 115 ms at 6‐month follow‐up (Table 1 and Figure S1).
Cardiac synchronization
At 1‐month follow‐up, echocardiographic signs of atrioventricular mechanical dyssynchrony demonstrated progressive alleviation. The pre‐ejection interval was shortened after PAB and CRT. Furthermore, parameters including anteroseptal‐post time to peak strain delay, septolateral time to peak strain delay, and longitudinal global strain all showed improvement (Table 1, Figure 3, and Figure S1).
Further follow‐up
Among the five patients who underwent BiVP‐CRT implantation, the procedure was successful in four. The remaining patient was converted to left bundle branch pacing (LBBP)‐CRT at the 1‐month follow‐up due to RV dyssynchrony. Additionally, two patients who had achieved normal cardiac function underwent transcatheter balloon dilatation 6 months after PAB surgery to relieve a high PAB pressure gradient.
Complications
Three patients developed nausea, vomiting, and abdominal distension. These symptoms were relieved at the 3‐month follow‐up and had completely disappeared at the 6‐month follow‐up. None of the patients experienced wound infection or pacemaker pocket exudation.
DISCUSSION
CRT is considered an effective therapy for improving symptoms and prognosis in adult patients with dysynchronous HF.13, 14 Previous studies have shown that CRT could benefit adults with moderate‐to‐severe chronic HF or congenital heart disease by improving ventricular function.15, 16, 17 However, these findings were not directly applicable to pediatric or infant populations, particularly regarding CRT for infants with LBBB‐associated DCM. Most current CRT research has focused on congenital heart disease in infants with prolonged QRS duration.18, 19, 20, 21, 22 CRT was utilized in only 16% (16/103) of reported infant DCM cases, including only one patient aged 3 months.20 It remains unknown whether CRT can lead to sustained hemodynamic improvement in infants with advanced HF and LBBB. Considering the prolonged recovery and limited reverse remodeling with pacemaker implantation alone, we propose a new strategy: CRT combined with PAB.
As a new indication for an established technique, PAB is used to treat end‐stage LV‐DCM with preserved RV function.10 Some infants and young children undergoing PAB have shown clinical benefits, including reduced LV volume and improved cardiac function.23, 24 The objective of this study was to investigate whether combined CRT and PAB therapy could improve cardiac output and synchrony in advanced HF infants with LV DCM and LBBB.
Therefore, five patients with DCM and LBBB were enrolled in this study. All patients presented with severe congestive cardiac dysfunction at admission. After CRT and PAB surgery, LVEDd, LVEDd z‐score, LVESd, and LVESd z‐score began to decrease by postoperative day 3, and changed significantly at 3 months.
The NT‐proBNP level increased transiently on day 3, likely due to surgical trauma, and subsequently decreased by 3 months. As LV volume decreased, cardiac function recovered, and ventricular synchrony was further improved. After 6 months of follow‐up, all patients exhibited considerable improvement or recovery in LVEF compared to baseline, accompanied by reductions in LVEDd, LVEDd z‐score, LVESd, LVESd z‐score, and NT‐proBNP.
No malignant arrhythmia or sudden cardiac death occurred during the follow‐up period. Our data consistently indicated that the combined CRT and PAB approach led to favorable short‐term outcomes in infants with LBBB‐associated DCM. Li et al.25 summarized outcomes in nine children (mean age, 3.0 ± 2.6 years) with cardiac dysfunction and complete LBBB (CLBBB).
All patients recovered to normal cardiac function 1 year postoperatively. Cecchin et al.19 reported on a cohort of 14 DCM children with wide QRS and cardiac dysfunction. After a median follow‐up of 0.7 years, the median QRS width decreased from 149 to 120 ms, and LVEF improved from 36% to 42%.
These studies suggested that CRT alone might lead to slow functional recovery, limited reverse remodeling, and modest functional improvement. In our study, key parameters demonstrated marked enhancement: median LVEDd decreased from 46 to 28 mm, NT‐proBNP fell from 3978 to 270 pg/mL, median QRS width narrowed from 138 to 115 ms, and LVEF improved from 26% to 65%. Compared with previous reports, our cohort of infants with advanced HF showed faster and more substantial recovery in cardiac function, with no adverse events observed.
Current research has indicated that PAB can be applied to children with DCM to improve cardiac function and promote reverse remodeling.23, 24 In our study, we enrolled infants younger than 1 year. The early postoperative PAB pressure gradient was 20–25 mmHg, which later increased to 40–45 mmHg at the 3‐month follow‐up and further to 50–55 mmHg at the 6‐month follow‐up. As the pulmonary artery pressure increased, the ventricular septum shifted toward the LV, resulting in a reduction of LV volume.
To prevent elevated PAB pressure gradient from impairing right heart function, and based on satisfactory cardiac function recovery, two patients underwent transcatheter balloon dilatation at 6 months after PAB surgery. This intervention aimed to maintain a deliberate residual gradient of 25–30 mmHg, thereby providing ongoing support for LV function. Schranz et al.23 reported that among their cohort, 42 patients (68.9%) showed cardiac improvement, and 34 (55.7%) achieved cardiac recovery.
The study demonstrated that PAB enables functional recovery and can serve as a bridge‐to‐recovery strategy for patients with end‐stage LV‐DCM. One clinically important finding of our study was the rapid and progressive improvement in ventricular function observed after combined CRT and PAB surgery. A possible explanation was that CRT corrects ventricular dyssynchrony, while PAB reduces ventricular volume; their synergistic action promotes rapid cardiac functional recovery.
Existing literature indicates that no infant with LBBB‐associated DCM has undergone PAB surgery alone. These infants typically present not only with severe LV dilation but also with significant interventricular dyssynchrony. PAB primarily functions by increasing RV afterload to alter the position and motion of the interventricular septum, thereby improving interventricular synergy.
However, its capacity to correct pre‐existing, significant intraventricular dyssynchrony is likely limited and incomplete. We therefore posit that CRT is essential for optimizing this complex, mixed type of mechanical dyssynchrony.
In our cohort, four patients received BiVP‐CRT, while one initially assigned to BiVP‐CRT in the early postoperative period was later crossed over to LBBP‐CRT due to inadequate cardiac synchronization. BiVP‐CRT is regarded as a standard treatment for children with dyssynchronous HF.5 Our study demonstrated cardiac recovery in all five patients treated with BiVP‐CRT, consistent with previous reports indicating that HF patients can benefit from BiVP‐CRT with functional improvement and symptom relief.12, 13, 14 One randomized controlled trial in adults further suggested that LBBP‐CRT led to greater improvement in LVEF and better mechanical resynchronization than BiVP‐CRT in patients with moderate HF and CLBBB.15 However, for pediatric HF patients, it remains unclear whether BiVP‐CRT or LBBP‐CRT is more conducive to cardiac recovery. The role of BiVP‐CRT in improving outcomes in infants with end‐stage LV‐DCM warrants further investigation with larger cohorts and multifactorial analysis.
Nevertheless, our results provide encouraging evidence that BiVP‐CRT can yield clinical benefits. Complications associated with CRT and PAB in infants are not yet fully understood. In the present study, three patients experienced nausea, vomiting, and abdominal distension, all of which resolved by the 6‐month follow‐up.
Multi‐site pacing and the potential proarrhythmic risk in such young patients deserve clinical attention.
The neonatal heart exhibits a remarkable capacity for myocardial regeneration, primarily driven by highly active cardiac progenitor cells.26 However, this regenerative potential is age‐dependent and markedly diminishes after the first year of life.27 The timing of intervention—particularly patient age—proved to be a key determinant of clinical outcomes. Previous studies, as well as our own institutional criteria, have emphasized the importance of age considerations in patient selection for PAB.23, 24, 28 In our experience, the procedure achieved optimal outcomes when performed before 1 year of age, whereas older age was associated with a poorer prognosis. FLNC and TTN gene variants are known to underlie cardiac and muscular phenotypic manifestations.
DCM associated with FLNC or TTN mutations often follows a malignant clinical course, characterized by a high risk of sudden cardiac death.29, 30, 31, 32 Infants with FLNC‐ or TTN‐related DCM typically present with aggressive HF and show limited response to guideline‐directed medical therapy. In such high‐risk cases, the combined use of CRT and PAB emerges as a promising therapeutic strategy to improve clinical outcomes and potentially avoid the need for heart transplantation.
This study has several important limitations. First, although the cohort is small, with only five patients treated at a single institution, this reflects the rarity and complexity of the condition and underscores the value of reporting these preliminary findings to inform future clinical strategies. Second, the follow‐up duration was relatively short, and the long‐term postoperative complication profile remains unclear.
Third, all patients presented with concurrent HF and LBBB at disease onset, making the causal relationship between cardiomyopathy and LBBB difficult to determine. Fourth, whether the combination of PAB and CRT can accelerate cardiac functional recovery and reverse remodeling compared to CRT alone requires further validation.
In conclusion, our study provides significant insights into the management of LBBB‐associated DCM in infants. The combined strategy of CRT and PAB demonstrates promising efficacy and feasibility, representing a potential advance in pediatric cardiology. Although these findings are preliminary, they underscore the procedural safety and clinical value of dual resynchronization therapy in young patients with advanced DCM.
Future multicenter collaborative studies or large‐scale registries are warranted to further validate the effectiveness and safety of this approach.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
Supporting information