Risk factors for development of ventricular tachycardia in patients with ventricular premature contraction with a structurally normal heart

Abstract Background We examined risk factors for development of ventricular tachycardia (VT) in pediatric patients with ventricular premature contractions (VPCs) and a structurally normal heart. Methods The subjects were 81 844 first graders and 88 244 seventh graders of Kagoshima City School‐based cardiovascular screening (SCV‐screening) between 2001 and 2015. We retrospectively reviewed the clinical data of students who were diagnosed as having VPC. Results Ventricular premature contractions were observed in 134 first graders (0.16%) and 270 seventh graders (0.31%). On the screening electrocardiograms (ECGs), 43 patients (11%) showed bi‐/trigemini, three patients (0.7%) showed a couplet, and one patient showed VT. We obtained 166 patients' follow‐up information and evaluated 59 patients (36%) as improved, 97 patients (58%) as no change, and 10 patients (6%) as worsened (couplets, five; triplets, two; VT, three). We assumed that these worsened patients have risk factors for development of VT. Comparing the findings of SCV‐screening ECGs of risk patients with the others, a significant difference was observed only in the number of VPCs (per 10 seconds) (mean ± SD; 4.3 ± 2.6 vs 1.8 ± 1.4, P < .0001). A logistic regression analysis revealed that the number of VPCs was significant (P < .001, odds ratio; 2.01, 95% confidence intervals; 1.46‐2.93). Receiver operating characteristics analysis showed an adequate cut‐off number of three VPCs for the risk, the sensitivity was 89% and the specificity was 77%. Conclusions Of the patients with VPC and a structurally normal heart, a few patients developed VT. Careful observation is important in patients who had three or more VPCs on SCV‐screening ECG.


| BACKG ROU N D
Ventricular premature contractions (VPCs) are commonly observed, not only in structural heart diseases, but also in structurally normal heart. Although the prognosis of VPC is generally considered good in a child with a structurally normal heart, 1 a small number of patients may experience worsened arrhythmia like ventricular tachycardia (VT). The risk of VT in patients with VPC with a structurally normal heart is not fully determined, and the incidence of those VPC patients in general pediatric population, which must be a basic information concerning its risk, is not also well-known. For adequate management of the patients with VPCs and to decrease parents' anxiety about their children, it is necessary to clarify the incidence of VPC with a structurally normal heart and to examine their risk factors for development of VT.
A nationwide school-based cardiovascular screening (SCVscreening) program for heart disease is implemented for all students of first, seventh, and tenth graders in Japan. 2 This screening system is useful for identifying high risk individuals among both athletes and nonathletes. [2][3][4] Using the data obtained from this program, we can assess the incidence and prognosis of VPC with a structurally normal heart in the general pediatric population.
The objective of this study was to examine the incidence and risk factors for development of VT in pediatric patients with a structurally normal heart, using VPC patients diagnosed in SCV-screening.

| ME THODS
In the Kagoshima City SCV-screening programs, students were examined at each school using 12-lead electrocardiograms (ECGs) for 10 seconds at a paper speed of 25 mm/s. Physical examination of all students by school doctors was also performed. Parents of the students were asked to submit a questionnaire concerning their children's heart diseases. With the information from the school doctors' examination, pediatric cardiologists read ECGs and identified students requiring a second examination. Using the information from the questionnaires, pediatric cardiologists identified students who had already been diagnosed with cardiac diseases or VPCs and were currently being followed-up by pediatric cardiologists at pediatric cardiac centers, and these patients were decided not to refer to the second examination.
During the second examination for patients with VPC, physical examinations by pediatric cardiologists, chest-radiography, and exercisestress ECG tests were performed. Ultrasound echocardiography was also performed, if necessary. After the second examination, patients with VPC were referred to pediatric cardiac centers for the follow-ups.
Subjects were the first and seventh graders who were diagnosed as having VPC with a structurally normal heart for the first time in SCV-screening between 2001 and 2015. Patients who had underlying heart disease were excluded from the study. In these VPC patients, we retrospectively reviewed the SCV-screening ECG and clinical data from the follow-up hospitals. We focused on the changes in presentation of VPCs during follow-up. The changes were defined as follows: (a) improved, patients showed disappearance of VPCs; (b) no change; (c) worsened, patients with no couplets at screening ECG developed couplets, triplets, or VT, or patients with a couplet at screening ECG increased number of couplets or VT. A VT was defined as occurrence of a series of three or more consecutive VPCs. 5 In this study, we used the term of triplet for only three consecutive VPCs. Patients of sudden cardiac death were also included in the patients with worsened change. We assumed that these worsened patients had risks of developing VT, and we examined the VPC patients diagnosed through SCV-screening. Patients with VPCs were monitored using Holter ECG and/or an exercise ECG (Master two step test and/or Treadmill ECG). Echocardiography was performed in patients when necessary. A follow-up period was defined as a period between SCV-screening and the last visit of pediatric cardiac center.

| Statistical analysis
Differences in heart rate and QRS duration between worsened and nonworsened groups were examined using a Mann-Whitney U test.
Differences in incidences between the two groups were examined using Fisher's exact probability test. Cut-off points were determined using receiver operating characteristic (ROC) curves to maximize the sensitivity and accuracy for discriminating patients who had risks of developing VT from the others. To examine the risk of patients, a logistic regression analysis was applied using the factors which were significantly different between patients with the risks and the others as the independent variables. All P values presented were two-sided, and values of P < .05 were considered statistically significant. Data were processed using a statistical program SPSS 25.0 (IBM Japan, Inc).

| Findings of the first screening ECG of SCVscreening
Among patients with VPC, there were slightly more female patients in the first graders and slightly more male patients in the seventh graders (Table 2). On SCV-screening ECGs, the number of VPCs observed was 1.8 ± 1.5 (mean ± SD) per 10 seconds in the first graders and 1.7 ± 1.5 per 10 seconds in the seventh graders. Most of the patients have one or two VPCs, those incidences were 76% in the first graders and 77% in the seventh graders ( Figure 1). Bigeminy or trigeminy was observed in 14 first graders (10%) and in 29 seventh graders (10%). All patients showed monofocal VPCs, and no patients showed multifocal VPCs. A couplet was observed in one first grader and two seventh graders, and a VT was observed in one first grader. The patterns of VPCs of first graders are different from those of seventh graders (Table 3): complete right bundle branch block (CRBBB) pattern VPCs were observed in 54% of the first graders with VPC but in 35% of the seventh graders (Table 3). Most VPCs showed an inferior axis in both CRBBB and complete left bundle branch block (CLBBB) pattern.

| Changes in presentation of VPC during followups
Among VPC patients diagnosed in the SCV-screening, we obtained 166 patients' follow-up information (proportion in the VPC patients: first graders, 44%; seventh graders, 39%) (median follow-up period: first graders, 3.1 years; seventh graders, 1.3 years). In the first graders, we evaluated 28 patients (47%) as improved, 28 patients (47%) as no change, and four patients (7%) as worsened ( the SCV-screening program.

| Factors of SCV-screening ECG related with risks for development of VT
We compared the findings of SCV-screening ECGs between the group of patients with risk for development of VT (high risk group) and the other groups (low risk group). Case J was screened by a school doctor and not by SCV-screening ECG. We, therefore, excluded this patient from the high risk group for this analysis, and found that no differences were observed in heart rate or in the QRS duration of VPC (Table 6). The incidence of CLBBB pattern of VPC was significantly higher in the high risk group of first graders than in low risk group (P = .034). However, this difference was not significant in the seventh graders. Significant differences were observed in the number of VPCs (per 10 seconds), and in the incidence of bi-/trigeminy in both first and seventh graders. The incidence of patients who had many VPCs (three VPCs or more per 10 seconds) was significantly higher in the high risk group than in the low risk group (first graders, 100% vs 29%, P = .011; seventh graders, 80% vs 19%, P = .008). The incidence of patients with a couplet was higher in high risk group. This difference was significant in the seventh graders (P = .002) but not in the first grad- and showed that the sensitivity was 75% and the specificity was 85%.
In the seventh graders, ROC analysis revealed an adequate cut-off number of VPCs for high risk patients (three VPCs per 10 seconds, AUC = 0.833) and showed that both the sensitivity and the specificity were 80%. Cut-off number of VPCs in all patients was three VPCs per 10 seconds (AUC = 0.863) and showed a sensitivity of 89% and specificity of 77%.

| D ISCUSS I ON
This study showed that the incidence of VPC patients with a structurally normal heart was 0.  available. Using SCV-screening ECGs, the incidence of VPCs was shown in several reports: 0.17%-0.32% in the first graders and 0.35%-0.50% in the seventh graders (Table 7). [9][10][11][12] Among these VPC patients, some patients had heart diseases. This study showed the lowest incidence among the previous reports, because we excluded patients with heart disease who had been already diagnosed before. SCV-screening ECGs are recorded only for 10 seconds, so that students diagnosed in the SCV-screening may have frequent VPCs (8640 or more VPCs per day, in a simple calculation). The PACES/HRS expert consensus statement showed that the management of patients with frequent VPCs, defined as more than 10% of beats in a 24-hour, should be followed longitudinally. 13 Considering this PACES/HRS guideline, SCV-screened VPC patients with a structurally normal heart require further examination.
Prognosis of these VPC patients with a structurally normal heart is thought to be benign and not to be associated with sudden, unexpected death. 1  This study showed that CRBBB with an inferior axis was the most frequent morphology of VPCs in the first graders and that CLBBB with an inferior axis was the most frequent morphology in the seventh graders. This difference in morphology between the first and seventh graders has been previously reported. 10 16 This report may explain the difference in morphology between the first and seventh graders. Still, the mechanism is not completely understood, it may be related to a developmental process of the conduction system. All high risk patients showed the CLBBB pattern and one first grader who had VT in the first screening ECG and improved after receiving medication showed the CRBBB pattern. These findings were interesting in considering prognoses for patients with VPCs identified by SCV-screening program.

| Limitations
This study has some limitations: a low percentage of follow-up students and a short follow-up period are two of them. In Kagoshima City, any cardiac event in a student, such as sudden death, out of hospital cardiac arrest, VT, or heart failure, results in them being transferred to one of the four pediatric cardiac centers. Since the pediatric cardiologists of the four pediatric cardiac centers are members of the committee of the SCV-screening, all students' information regarding the events are reported to the committee of the SCV-screening and every reported patient's SCV-screening ECG is rechecked. The committee did not receive any information of cardiac events of the VPC patients diagnosed by SCV-screening during the study period, suggesting that no events occurred in the VPC patients who dropped out from the follow-up in this study. The ratio of VPC patients with a favorable prognosis was larger than that shown in this study. Another limitation involved our using the ECGs at the first screening as the basement data to identify their prognosis. SCV-screening ECGs are recorded for only 10 seconds.
This short period of recording does not always reflect a patient's condition. Therefore, there might be a possibility of an underestimation of the basement data. Despite these limitations, the data shown in this study are important and useful for the pediatric cardiologists to make plans for further examination and follow-up when VPC patients are identified through the SCV-screening program and are referred to their hospitals.
The information is also valuable for the parents of the children with VPC to ensure that they receive proper follow-up after the initial screening.

| CON CLUS IONS
The incidence of patients with VPC and a structurally normal heart in the SCV-screening was 0.16% in the first graders and 0.31% in the seventh graders. VPC patients with a structurally normal heart diagnosed in SCV-screening commonly showed improved or no changes in ECG during the follow-ups, but a few patients developed VT.
Careful observation is important in patients who had three or more VPCs identified on the SVC-screening ECG.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interests for this article.