Although ventricular premature complexes (VPCs) have been shown to correlate with decreased cardiac function in adults, the correlation of left ventricular (LV) function to VPCs in asymptomatic children remains unclear. The aim of this study was to determine the correlation of VPC burden with LV function in asymptomatic pediatric patients with structurally normal hearts. This was a retrospective analysis of patients aged ≤21 years with echocardiograms and 24-hour Holter monitors with ≥0.5% VPCs completed within 60 days of each other. LV fractional shortening (FS) was compared with VPC burden and VPC characteristics. Normal LV function was defined as FS ≥28%. Correlation between VPC burden and LV function was determined by regression analysis. Wilcoxon’s rank-sum test was used to compare LV function with VPC characteristics. This study included 123 patients (77 male [63%]). The median age was 11.6 years (interquartile range 5.8 to 14.3). The median VPC burden was 11.2% (interquartile range 4.8% to 18.9%), and median FS was 36% (interquartile range 33% to 38%). There was no significant correlation between VPC burden and LV FS (p = 0.50). The presence of uniform versus multiform VPCs (p = 0.29), ventricular couplets (p = 0.37), or runs of ventricular ectopy (p = 0.19) were not associated with a decrease in LV FS. Twenty-two patients (18%) had VPC burden >24%, none of which had decreased LV FS. In conclusion, there was no significant relation between VPC burden or VPC characteristics and LV systolic function in this pediatric population with structurally normal hearts.
Although ventricular premature complexes (VPCs) may be regarded as a marker of underlying heart disease in adults, VPCs in children are historically perceived as benign in structurally normal hearts. VPC location and frequency can vary, the morphology can be uniform or multiform, and ventricular couplets and triplets often occur. Up to 40% of healthy children will have VPCs on routine Holter monitoring. To date, studies of long-term sequelae of VPCs in pediatric patients with structurally normal hearts involve small cohorts and are few in number. In adults, correlations between VPC burden and decreased left ventricular (LV) systolic function, as well as VPC characteristics and cardiomyopathy have been reported. Baman et al demonstrated that a VPC burden >24% was associated with decreased LV systolic function and that the LV systolic dysfunction was reversible in most of these patients after catheter ablation of the VPCs. Similarly, Niwano et al demonstrated that in asymptomatic adults with structurally normal hearts, there was a small but significant negative correlation between VPCs and LV systolic function. The relation between decreased cardiac function and VPC burden in children remains unknown. There have been small studies demonstrating decreased cardiac function in children with VPCs, but there have been no large reviews assessing this relation. The purpose of this study was to examine the relation between VPC burden on 24-hour Holter monitoring and LV systolic function on echocardiography in pediatric patients with structurally normal hearts.
Methods
This was a retrospective cohort study of pediatric patients at a single pediatric institution with institutional review board (ID number 2013-6912) approval. Patients aged 0 to 21 years who underwent 24-hour ambulatory electrocardiographic (Holter) monitoring from January 1998 to June 2013 that showed ≥0.5% VPCs were identified. Patients with structurally normal hearts on echocardiography performed within 60 days of the study Holter monitoring were included. Patients with the following minor abnormalities were included: mild valve regurgitation, mitral valve prolapse, bicuspid aortic valve, patent foramen ovale, and small atrial septal defect. Patients with known channelopathies, cardiomyopathies diagnosed before the initial Holter study, ventricular tachycardia, and/or congenital heart disease were excluded.
Holter monitoring was completed at the discretion of the ordering physician. If multiple Holter studies were completed for a patient during the study period, the first Holter study that showed a VPC burden ≥0.5% was identified and used for analysis. Holter monitors were placed for 24 hours with a standard 2-lead configuration (modified V 1 and V 5 ). VPC burden was defined as the percentage of total VPCs divided by the total number of ventricular (QRS) complexes. VPC characteristics evaluated included uniform, multiform (>1 VPC morphology), VPC morphology (right bundle vs left bundle branch block), coupling intervals (fixed vs variable), ventricular couplets, and runs of ventricular ectopy ≥3 beats. VPC morphology was determined in patients with uniform VPC morphology by 12-lead electrocardiography performed within 30 days of the study Holter monitoring. A pediatric electrophysiologist confirmed all Holter studies and VPC characteristics.
Echocardiography was performed at the discretion of the ordering physician and included in the analysis if completed within 60 days of the study Holter monitoring. The echocardiograms were reviewed for anatomy, qualitative LV systolic function, and LV fractional shortening (FS). LV diastolic dimension and FS were measured by M-mode echocardiography after 2 consecutive sinus beats. Normal LV systolic function was defined as FS ≥28%. A pediatric cardiologist confirmed all echocardiographic findings. A second pediatric cardiologist who was blinded to the study patients reviewed 10% of the echocardiograms for interobserver reliability.
Demographic data were obtained on all patients at the time of the study Holter monitoring. Demographic data included age, gender, height (centimeters), weight (kilograms), treatment with antiarrhythmic medication at the time of Holter monitoring, and history of catheter ablation of VPCs before Holter monitoring.
We identified patients with follow-up Holter studies within 6 to 36 months of the initial Holter studies. The first follow-up Holter recording within this time period was evaluated for VPC burden and compared with that on the initial study Holter recording. Similarly, the first follow-up echocardiogram obtained within 60 days of any follow-up Holter recording was evaluated for LV FS.
Statistical analysis was performed using Stata version 10.0 (StataCorp LP, College Station, Texas). Demographic and clinical characteristics of the patients were summarized using measures of central tendency, variability, and frequency. Medians and interquartile ranges (IQRs) are reported for continuous variables. The primary outcome variables were VPC burden and LV FS. Correlation between VPC burden and LV FS was determined by regression analysis. The correlation between LV FS and VPC characteristics and the change in VPC burden over time were determined by Wilcoxon’s rank-sum test. A p value <0.05 was considered statistically significant.
Results
Inclusion criteria were met by 123 patients, all of whom were included in analysis. There were 405 patients with Holter recordings that showed ≥0.5% VPCs, with 154 excluded for lack of echocardiographic examinations completed within 60 days of Holter monitoring and 33 excluded for poor echocardiographic images that precluded accurate FS measurement. Fifty-eight patients were excluded for complex congenital heart disease, 27 for previous diagnosis of ventricular tachycardia, and 2 for diagnosis of channelopathy. An additional 8 patients were excluded for previous diagnosis of cardiomyopathy: 5 with dilated cardiomyopathy, 2 with mitochondrial myopathy, and 1 with hypertrophic cardiomyopathy. Chart review was completed for all patients with cardiomyopathy to confirm that the diagnosis was not VPC-induced cardiomyopathy. Baseline demographic data of the 123 patients included for analysis are listed in Table 1 . The median VPC burden was 11% (IQR 5% to 19%). Of patients with corresponding electrocardiograms (n = 84), 58 (69%) demonstrated left bundle branch block morphology. The median LV FS was 36% (IQR 33% to 38%), with no patients demonstrating FS <28%. LV FS z scores were within 2 SDs (≥−2) for 122 patients (99%). There was no evidence of LV dilatation, with LV diastolic dimension z scores within 2 SDs (none ≥+2). No patients had undergone VPC catheter ablation, and 1 patient was treated with antiarrhythmic therapy (atenolol) before Holter monitoring. Most Holter recordings and echocardiograms (60%) were completed on the same day.
| Variable | n=123 |
|---|---|
| Age (years) | 12 (6 – 14) |
| Male | 77 (63%) |
| Height (cm) | 152 (122 – 167) |
| Weight (kg) | 41 (21 – 62) |
| Body Surface Area (m 2 ) | 1.33 (0.87 – 1.6) |
| Holter Characteristics | |
| Ventricular Premature Complex Burden (%) | 11 (5 – 19) |
| Couplet | 54 (44%) |
| Runs | 30 (24%) |
| Uniform | 99 (80%) |
| Variable Coupling | 96 (78%) |
| Right Bundle Branch Morphology | 30 (24%) |
| Left Ventricular Shortening Fraction (%) | 36 (33 – 38) |
| Days between Holter and Echocardiogram | 0 (0 – 57) |
| Antiarrhythmic Medications | 1 (0.8%) |
| Prior Catheter Ablation of Ventricular Premature Complexes | 0 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
