Introduction
Pediatric vasovagal syncope (VVS) in children is the most common hemodynamic type of neurally mediated syncope (NMS), which manifests as syncope or presyncope (e.g., dizziness, headache, chest tightness, chest pain, hyperventilation, amaurosis, etc.) during prolonged standing, or a sudden change of body position (e.g., from supine or sitting or squatting to an upright position), or in a stuffy environment [1-4]; it is termed malignant VVS if it manifests itself as recurrent episodes accompanied by a cardiac arrest lasting for more than 3 seconds during a syncope episode, and it also could be triggered by the head-up tilt test (HUTT) or by stimuli such as emotion or pain [5]. HUTT induces a vasovagal response, resulting in a decrease in heart rate (HR) and blood pressure (BP), and even cardiac arrest, which can clinically lead to sudden cardiac death [6]. The probability of HUTT-induced sinus arrest is about 3.5%-4.7% [7,8]. The presence of cardiac arrest as an independent risk factor for severe clinical manifestations such as seizure-like activities during HUTT-induced syncope [9]. Malignant VVS is gaining clinical attention because of its possible serious risk of death.
There are relatively few relevant literature on pediatric malignant VVS. Children with malignant VVS tend to have their syncopal episode at an earlier age, and showed severe autonomic nerve dysfunction, characterized by the dominance of the parasympathetic nerve [10]. Some researchers have found that the mean red blood cell hemoglobin and the standard deviation of the RR interval on the electrocardiogram (ECG) are independent risk factors for pediatric malignant VVS [5]. The average age, height, weight, and body mass index (BMI) levels are lower, and sex has no significant impact [7].
This research conducted a retrospective analysis of 24 years of clinical data on pediatric VVS at a single center, aiming to explore the demographic characteristics and independent influencing factors of pediatric malignant VVS, and to conduct linear and nonlinear analyses. The goal was to assist HUTT examiners in earlier identification or prevention of cardiac arrest, reduce the harm caused by HUTT-induced malignant VVS cardiac arrest to children, and explore non-invasive, economical and effective predictive methods for predicting malignant VVS, create greater value for patients and examiners.
Methods
1. Research population
Data of 3,734 children, aged 3–18 years old, 1,659 males and 2,075 females, first diagnosed with VVS due to unexplained syncope or presyncope in Department of Pediatric Cardiovasology of The Second Xiangya Hospital, Central South University, were retrospectively analyzed from September 2000 to January 2024. The causes of syncope or presyncope remained unclear after detailed history inquiry, careful physical examination, 12-lead ECG, Holter ECG, cardiac radiograph, echocardiogram, electroencephalogram, cranial computed tomography or magnetic resonance imaging, and blood biochemistry examination (including fasting glucose and cardiac enzymes) to exclude organic cardiac, cerebral, pulmonary, and vascular disorders, immune diseases, and cardiac causative disorders, etc. HUTT was performed after obtaining written informed consent from the patients or/and legal guardians. 783 children with VVS who showed HR slowing during HUTT were collected and divided into malignant VVS group (122 cases) and nonmalignant VVS group (661 cases) (Fig. 1).
The research was conducted according to the guidelines of the Declaration of Helsinki, and approved by The Second Xiangya Hospital, Central South University (Ethical Audit No. Study 249(2022)).
2. HUTT
1) Preparation before the test
Subjects stopped using cardiovascular active drugs for more than 5 half-lives before the test, and stopped using foods that may affect autonomic function such as coffee. Fasting and abstaining from food and drink 4 hours before the test. Introduce the pretest precautions and possible risks during the test to the subject or guardian, and the written informed consent was signed by the subject or legal guardian. HUTT was divided into basic HUTT (BHUT) and sublingual nitroglycerin-activated HUTT (SNHUT).
2) Basic HUTT
The tilt devices were ST-711 Head-up Tilt Test System (Beijing Juchi Pharmaceutical Technology Co., Ltd., China) and SHUT-100 Head-up Tilt Test System (Jiangsu Standard Medical Technology Co., Ltd., China). The examination time was chosen from 8:00 AM to 11:00 AM, the room temperature was 20°C –24°C, and in a quiet environment, the subjects emptied their bladders, rested supine on the tilt diagnostic bed for 10 minutes, the chest and knee joints were secured with a harness, and the baseline HR, BP, and ECG were recorded, and then the subjects were converted to the head-high, foot-low, and tilt position of 60° in 15 seconds, and the HR, BP, ECG and clinical manifestations were continuously monitored and recorded until a positive reaction occurred. BP was monitored by inflatable cuff. The test was terminated after a positive reaction and returned to the horizontal position within 10 seconds. The patients were monitored at any time if symptoms of discomfort with syncope and presyncope appeared. In case of discomfort symptoms were monitored at any time until a positive reaction occurred or the 45 minutes specified in the test procedure was reached [1].
3) SNHUT
Those with negative BHUT continued to maintain the same tilt angle position [1], subjects were given sublingual nitroglycerin tablets 4–6 μg/kg (maximum amount ≤300 μg), and HR, BP, ECG and clinical manifestations were continuously monitored until a positive reaction appeared or the test was terminated if there was no positive reaction for a continuous period of 20 minutes, tilted diagnostic beds were rapidly returned to the horizontal position, and monitoring of HR, BP, ECG and clinical manifestations were continued after the supine position until the above indexes returned to normal.
4. Diagnostic criteria for VVS
Syncopal episodes or presyncope symptoms together with any of the following responses in the HUTT are considered positive responses [1,11]: (1) systolic BP (SBP) ≤80 mmHg (1 mmHg=0.133 kPa) or diastolic BP (DBP) ≤50 mmHg or mean pressure decrease ≥25%; (2) HR <75 bpm for 3 to 6 years old children, <65 bpm for >6 to 8 years old children, and <60 bpm for those >8 years children and adolescents; (3) ECG showing sinus arrest, or junctional escape rhythm; (4) atrioventricular block (II or III degree) or cardiac arrest ≥3 seconds.
According to the guidelines [1], a cardiac arrest lasting longer than 3 seconds during a syncopal episode or during the HUTT is defined as malignant VVS.
5. Variables
Age, sex, height, weight, BMI, times of syncopal episodes in medical history, HR and BP (SBP, DBP) at baseline and at the time of HUTT-induced syncope, HUTT modes (BHUT, SNHUT).
6. Statistical analysis
The continuous variable which normally distributed was expressed as mean±standard deviation and nonnormal distribution was expressed as median (interquartile range). Categorical variables were expressed in frequency or as a percentage. χ2 (categorical variables) and Student t test (normal distribution) compared between groups. Multiple logistic regression to analyze the possible association between potential influencing variables and malignant VVS. Smooth curve fitting was employed to analyze threshold saturation effects, constructing regression models while adjusting for confounding factors. The likelihood ratio test validated nonlinear relationships, with curves plotted to identify threshold points. Segmented regression verified differences in effects before and after the threshold. All the analyses were performed with the statistical software Packages R ver. 4.4.1 (http://www.R-project.org, The R Foundation) and EmpowerStats (http://www.empower stats.com, X & Y Solutions, Inc., USA). P values<0.05 (2-sided) were considered statistically significant.
Results
1. Comparison of general information
Age, height, weight, BMI, and HR, SBP, and DBP at the time of syncope were decreased in the malignant VVS group compared to the nonmalignant VVS group (P<0.01), and the differences in sex, baseline HR and BP, and HUTT modes were not statistically significant between the 2 groups (P>0.05) (Table 1).
2. Univariate analysis for malignant VVS
Univariate analysis showed that age, height, weight and BMI were potentially protective against malignant VVS (all P<0.05). HR, SBP, and DBP during HUTT-induced syncopal episodes may have an inverse correlation with the occurrence of malignant VVS (all P<0.05) (Table 2).
3. Multifactorial analysis of malignant VVS
Multifactorial analysis showed that by adjusting for sex, times of syncopal episodes, HR, SBP, DBP, and HUTT modes at baseline and during syncopal episodes, it was found that age and BMI remained stable (odds ratio [OR] fluctuation <5%, P<0.05) compared with the results of the univariate analysis, suggesting that there were independent protective effects against malignant VVS. That is, for every 1-year increase in age and 1 kg/m2 increase in BMI, the risk of malignant VVS occurrence was reduced by 12% and 9%, respectively (Table 3). Subgroup analysis of the HUTT model revealed that age was an independent protective factor for pediatric malignant VVS who tested positive for SNHUT, while BMI served as an independent protective factor pediatric malignant VVS who tested positive for BHUT. Given the consistent direction of OR effect across subgroups, the lack of statistical significance for these factors in different HUTT model subgroups may be attributed to sample size limitations.
4. Multicollinearity analysis of age and BMI in malignant VVS
Age and BMI are highly correlated variables in the growth and development of children, necessitating multicollinearity testing to assess the reliability of conclusions. After statistical calculation, the variance inflation factor index for both age and BMI was 1.3, suggesting only minor multicollinearity. Therefore, no special handling is required, and these 2 variables do not significantly impact the conclusions.
5. Smooth curve fitting and saturation effect analysis for malignant VVS
Fig. 2 and Table 4 show that age and BMI were nonlinearly associated with malignant VVS. The saturation effect of age and change in risk of malignant VVS occurred at 12.9, i.e., for age < 12.9 years, the risk of malignant VVS decreased by 20% for every 1-year increase in age, and for age ≥ 12.9 years, the risk of malignant VVS was not significant for every 1-year increase in age. However, the nonlinear impact of BMI on malignant VVS is not significant.
6. Smooth curve fitting and saturation effect analysis for malignant VVS
The ability of the autonomic nervous system to self-regulate changes with age. We divided the age range into 4 equal parts and observed the changes in various indicators as age increased (Table 5). As shown in section 3.4, the incidence of malignant VVS gradually decreases and slows down with increasing age, and the trend is consistent with that of Fig. 2.
Discussion
HUTT is a common non-invasive method for assessing autonomic function, which can induce convulsions or arrhythmias [1,12-14]. Studies report that arrhythmias like sinus bradycardia, junctional escape rhythm, and sinus arrest are common following a positive HUTT response [13,14]. For instance, Wang et al. [13] observed convulsive syncope during HUTT in 28.09% of patients, with sinus arrest lasting 3.00–14.60 seconds [13]. The occurrence of sinus arrest appears independent of age, sex, or HUTT protocol. Notably, the timing of arrhythmias differs between the baseline phase (occurring concurrently with syncope/presyncope and BP decrease) and the sublingual nitroglycerin phase (often occurring after syncopal symptoms and BP decrease), potentially due to pediatric sympathetic dominance, drug effects, and other factors [13,14].
The incidence of malignant VVS in our research was 3.3% (122/3,734 cases), which is consistent with the 3.5%–4.7% results from of Sun et al. [7] and Numan et al. [8]. However, the incidence of HUTT-induced cardioinhibitory reaction with cardiac arrest in adults was 6.6% [15]. The incidence rate of malignant VVS induced by HUTT in children and adults is inconsistent, which is considered to be related to the following factors [1,16-19]: (1) Individual differences: children and adults differ in basic health, cardiovascular function, and autonomic nervous function. (2)Test methods/conditions: inconsistent tilt angles (usually 60° for children, 70°–80° for adults) and durations across studies lead to varying incidence. (3) Pretest preparation: patients’ diet, fluid intake, and rest status before HUTT may impact results. (4) Inconsistent diagnostic criteria: VVS diagnostic standards vary between studies. (5) Environmental factors: ambient temperature, humidity, and noise during testing affect patients’ psychological/physiological states, altering syncope incidence. (6) Drug effects: pretest use of certain drugs influences cardiovascular responses, causing different syncope rates. The average age of episode for pediatric malignant VVS is 10.86 years [10], slightly higher than the median age of 9 years reported by Sun et al. [7]. Among school-age children with VVS, they are prone to being induced into sinus arrest (referred to as "sinus arrest") during HUTT, which may be related to the fact that the pediatric autonomic nervous system in this age group is still in the developmental stage and has unstable autonomic function. The research by Brignole et al. [20] showed that in adult patients with syncope over 40 years old, the incidence of spontaneous cardiac arrest-induced reflex syncope monitored by implantable loop recorders was not associated with age; however, the incidence of cardiac arrest-induced syncope by HUTT was higher in younger patients and showed a linear decreasing trend with age. Additionally, there is a key feature in the HUTT examination: the smaller the magnitude of BP decrease, the greater the possibility of HR reduction and cardiac arrest. This reaction conforms to the 2B type manifestation in the International Study Standard for Vasovagal Syncope.
In this research, age showed an independent protective effect on malignant VVS. However, it could only significantly reduce the risk of malignant VVS in individuals younger than 12.9 years old. Once the saturation point was reached, the protective effect no longer became significant. This result is inconsistent with previous studies. Wang et al. [21] reported a positive correlation between age and pediatric VVS syncope symptoms, with the effect concentrated in those over 10.67 years old. The discrepancy stems from different research focuses: this study targeted HUTT-induced malignant VVS, whereas Wang et al. [21] focused on naturally occurring clinical syncope. Young children with immature autonomic nervous function are less sensitive to HUTT stimulation, making autonomic imbalance less likely than in natural syncope scenarios. Before the age of 12 years, the body is in a critical stage of the maturation of autonomic nerve and vascular functions. As the child grows older, the vascular elasticity gradually transitions towards that of adolescents (but not reaching the easily expandable state), the regulatory ability of the autonomic nerve over BP and HR increases, and the probability of excessive activation of the vagus reflex (such as inducing sinus arrest) decreases; at the same time, with the increase in age during this period, the body's tolerance to hypovolemia and metabolic fluctuations also improves, and these are important triggers for pediatric malignant VVS. Therefore, the risk decreases with age. After the age of 12.9 years, the human autonomic nerve development is basically mature, and the vascular regulatory mechanism becomes stable – even if there is a possible dilation of the inferior vena cava (a characteristic related to adolescent syncope), the mature autonomic nerve can balance the vascular response and avoid malignant events caused by abnormal reflexes. Therefore, further age growth will not significantly change the risk of malignant VVS [22].
In this research, BMI has a protective effect on pediatric malignant VVS, which is consistent with the conclusion reported by Yamada and Yanagimoto [23] that children with low BMI are more likely to experience malignant VVS (such as sinus arrest) during HUTT; low BMI may increase the probability of malignant VVS due to insufficient circulating blood volume, poor venous compliance (which can lead to a sudden decrease in return blood volume), and autonomic nerve imbalance (with the vagus nerve being dominant). BMI, a composite index influenced by height and weight, exerts a protective effect on pediatric malignant VVS. Specifically, higher height increases hydrostatic pressure, while greater weight enhances the muscle pump function of the abdomen and lower limbs; the combined effects of these 2 factors may underlie BMI's protective role. It is speculated that the combined factors of height and weight may be related to the slow response to pediatric syncope, and the specific mechanism remains to be further studied.
The duration of sinus arrest induced by BHUT in pediatric malignant VVS ranged from 3 to 20 seconds, with a mean duration of 5.87±2.92 seconds. Among these, 27.87% (34 of 122 cases) of cardiac arrests occurred during the BHUT phase, while 72.13% (88 of 122 cases) occurred during the SNHUT phase. The latter phase accounted for a higher proportion, which is associated with excessive vagal stimulation following lower limb congestion exacerbated by nitroglycerin administration [24]. In malignant VVS, the number of BHUT cases accounted for 12.64% (34 of 269) of the total BHUT cases, and the number of SNHUT cases accounted for 17.12% (88 of 514) of the total SNHUT cases. The difference in the proportion between the two is relatively small, so the error caused by the difference in total number can be disregarded. Before and after cardiac arrest, the activity of the vagus nerve increases sharply, and HUTT causes a significant decrease in cerebral perfusion. Excessive vagus nerve activity at rest may also prolong the duration of cardiac arrest [8,24-26]; In adult HUTT, a 60° tilt induces cardiac arrest later than an 80° tilt, but does not affect the duration of the arrest [27]; Reduced sympathetic activity and abnormal myocardial adrenergic innervation are key factors in some unexplained cardioinhibitory VVS cases [28]. Even with sustained sympathetic activity, presyncope may involve decreased maximum myocardial contractility and bradycardia, followed by complete loss of muscular sympathetic nerve activity and subsequent cardiac arrest [29]. In systolic vasomotor vagal responses, enhanced left ventricular contractility triggers presyncope (associated with increased vagal and sympathetic activity), while a sudden drop in sympathetic activity directly induces cardiac arrest and peripheral vasodilation.
Literature consistently indicates favorable prognosis for malignant VVS. For children, VVS patients showed no poor outcomes during follow-up, even when HUTT-induced seizure-like activities [9]. In adult studies: Guaraldi et al. [30] (1996–2010) found that for cardioinhibitory VVS patients with HUTT-induced sinus arrest (>35 seconds), syncope recurrence was common but cardiac arrest did not mean death or poor prognosis. Barón-Esquivias et al. [27] analyzed the largest adult cohort with HUTT-induced cardiac arrest, noting no association between cardiac arrest and malignant prognosis, nor did treatment (pacemaker/pharmacology) or clinical factors (sex, age, arrest times/duration) affect syncope frequency; Sau et al. [31] (1998–2012; 5,133 adults with HUTT-induced cardiac arrest >15 seconds) confirmed prolonged arrest did not guarantee poor prognosis, with most patients improving spontaneously long-term. Takase et al. [32] pointed out that the impairment of autonomic nerve function solely related to upright stress plays a significant role in the pathogenesis of cardiac arrest induced by HUTT; while the impairment of autonomic nerve function in the daily state has little impact on NMS patients who experience cardiac arrest induced by HUTT. Moreover, ganglion plexus catheter ablation can improve the prognosis of pediatric malignant VVS: a 2.5-year follow-up (0.6–5 years) showed that the postoperative child's syncope symptom score (3 [2–4] points vs. the preoperative 5 [3–8] points, P<0.01) was significantly reduced, 85% (17 of 20 cases) no longer had syncope, 80% (16 of 20 cases) had HUTT become negative after the operation, and there were no adverse prognoses such as arrhythmia [33]. Induced cardiac arrest time in HUTT is a poorly reproducible phenomenon, and implantation of a permanent pacemaker should not be used as a first-line treatment option for malignant VVS [34].
Although this study identified the independent protective effects and nonlinear relationships of age and BMI on pediatric malignant VVS, several limitations should be noted. First, HUTT only simulates triggers via BHUT or SNHUT, failing to replicate natural triggers (e.g., emotional stress, pain, prolonged standing); thus, conclusions are limited to HUTT-induced malignant VVS and not directly generalizable to naturally occurring cases. Second, potential biases exist in sample grouping and HUTT modes: while BHUT/SNHUT proportions were balanced between the malignant (122 cases) and nonmalignant (661 cases) VVS groups, SNHUT may be more prone to inducing malignant VVS due to nitroglycerin’s effects on lower extremity congestion and vagal activity. Insufficient adjustment for mechanistic differences between modes, coupled with a relatively small sample (783 cases, 20.97% of the initial 3,734 VVS patients), may introduce selection bias. Third, the retrospective single-center design (24-year data from one center) limits validity: reliance on historical clinical data may lead to collection biases (e.g., incomplete syncope details, missing autonomic function indicators like HR variability, or lack of follow-up), precluding analysis of age/BMI-related long-term prognosis. Additionally, single-center samples with fixed geographic and clinical characteristics restrict external generalizability across regions or medical resource levels. Finally, conclusions are only applicable to children with cardio-inhibited or mixed-type VVS, not all pediatric VVS cases.
In conclusion, age and BMI are independent protective factors for malignant VVS, and age is nonlinearly correlated with malignant VVS. Before the age of 12.9, the incidence of malignant VVS gradually decreases with the increase in age, and thereafter there is no significant impact. Demographic factors such as age and BMI can to some extent assist the HUTT inspectors in identifying or preparing for cardiac arrest earlier, and to reduce the damage caused to the children by the occurrence of malignant VVS cardiac arrest due to HUTT.
PDF Links
PubReader
ePub Link
XML Download
PubMed
Download Citation
