An analysis of vascular properties using pulse wave analysis in patients with vasovagal syncope

Abstract Background Vasovagal syncope (VVS) is a common cause of recurrent syncope. Nevertheless, the exact hemodynamic mechanism has not been elucidated. Pulse wave analysis (PWA) is widely used to evaluate vascular properties, as it reflects the condition of the entire arterial system. Hypothesis Cardiovascular autonomic modulation may influence the hemodynamic mechanism and result in different vascular properties between VVS patients and healthy individuals. Methods We enrolled consecutive patients diagnosed with VVS on head‐up tilt testing from January 2014 to August 2019. Healthy subjects were enrolled as the control group. We performed PWA on all participants. Using propensity score matching, we assembled a study population with similar baseline characteristics and compared hemodynamic parameters. Results A total of 111 VVS patients (43 ± 18 years, 72 females) and 475 healthy control subjects (48 ± 13 years, 192 females) were enrolled. Compared to the healthy control subjects, the VVS patients had a higher augmentation index (AIx) adjusted to a heart rate of 75 beats per minute (AIx@HR75, 20.5 ± 13.1% vs 16.7 ± 11.9%, P = .003). After 1:1 matched comparison (111 matched control), VVS patients consistently showed higher AIx@HR75 (20.5 ± 13.1% vs 16.7 ± 12.9%, P = .02) than the matched control group. According to age distribution, VVS patients showed significantly higher AIx@HR75 (10.6 ± 11.7% vs 2.5 ± 11.1%, P = .01) in a young age (15‐33 years) group. Conclusions VVS patients had greater arterial stiffness than healthy subjects. This is one of the plausible mechanisms of the pathophysiology of VVS.


| INTRODUCTION
Syncope is defined as transient loss of consciousness due to cerebral hypoperfusion, characterized by rapid onset, short duration, and complete spontaneous recovery. 1,2 According to recent guidelines, syncope can be divided into three main groups: reflex, cardiovascular, and secondary to orthostatic hypotension. Vasovagal syncope (VVS), mediated by the vasovagal reflex, is the most common presentation of syncope in the general population. 1,3 Recently, the consensus precise pathophysiological mechanisms underlying VVS suggested that the autonomic nervous system is the final common pathway leading to syncope. Therefore, cardiovascular autonomic modulation may play a role in the occurrence of syncope. [3][4][5] Nevertheless, the exact hemodynamic mechanisms and the relationship with autonomic regulation has not been elucidated. 6,7 Arterial pulse wave analysis (PWA) is a noninvasive index of arterial distensibility now generally advocated for the assessment of cardiovascular risk as well as for measuring blood pressure. It reflects central and peripheral vascular properties by measuring arterial stiffness and elastic compliance. [8][9][10] Therefore, we investigated central hemodynamics using PWA and compared VVS patients with healthy control subjects. We hypothesized that cardiovascular autonomic modulation may influence the hemodynamic mechanism and that vascular properties differed between VVS patients and healthy individuals.

| Study design
We enrolled 133 consecutive patients diagnosed with VVS at our institution from January 2014 to August 2019. The diagnosis of VVS was made using the head-up tilt (HUT) test based on the current diagnostic guidelines. 1,2 A positive response is defined as inducible presyncope or syncope associated with hypotension, with or without bradycardia (less commonly asystole). All patients were free from medication that could influence vascular properties and autonomic nervous system, including antihypertensive and neuromuscular drugs. We excluded patients with conditions that can affect vascular properties and hemodynamics (eg, hypertension, diabetes mellitus, renal disease, cerebrovascular disease, coronary or peripheral vascular disease, and structural heart disease) and those with arrhythmias and psychiatric disorders. We defined and classified VVS based on the modified Vasovagal Syncope International Study criteria as follows: type I, mixed; type II, cardio-inhibitory; type III, vasodepressor. 11 As the control group, we enrolled healthy subjects who were free from any syncope or presyncope, or who showed negative HUT test.
Additionally, subjects without VVS (eg, postural orthostatic tachycardia syndrome and orthostatic intolerance without tachycardia) on the HUT test were included in the control group.
The study design was approved by the Institutional Review Board of Inha University Hospital, Incheon, South Korea and was conducted in compliance with the ethical principles outlined in the Declaration of Helsinki (INHAUH 2019-08-012).

| HUT test
The tilt table test was performed with the patient in fasting state for 2 to 4 hours before the test was conducted in a quiet, closed room according to the recent standardized protocol. The patient was stabilized in the supine position for 5 minutes without venous cannulation and for 20 minutes with venous cannulation. The tilted angle was maintained between 60 and 70 for 20 minutes to induce syncope. If syncope was not induced, we started isoproterenol challenge at an incremental infusion rate from 1 to 3 μg/minute to increase the average heart rate by about 20% to 25% over baseline. Twelve-lead electrocardiogram (ECG) tracings and BP were measured every 2.5 minutes. The test was continued until the development of positive signs or completion of the protocol.

| Pulse wave analysis
To investigate vascular properties, we used the SphygmoCor (AtCor Medical Pty Ltd Head Office, West Ryde, Australia) system in all subjects. The examination was performed in a quiet room with a comfortable room temperature and the patient in a supine position.
The carotid and femoral pulse waves were analyzed, estimating the delay in theECG wave and calculating the pulse wave velocity (PWV). In addition to the estimation of radial and central blood pressure, central hemodynamic parameters including ejection duration (ED), the time to the peak/shoulder of the first (T1), and second pressure wave components (T2) during systole, the time to return of the reflected pressure (Tr) wave, P1 height, aortic pulse pressure (PP), augmentation pressure (AP), augmentation index (AIx), subendocardial viability ratio (SEVR) were estimated from the aortic wave morphology. 12,13 The P1 height is defined as the difference between the central pressure at T1 and the diastolic pressure. The AIx was defined as the augmented pressure (magnitude of wave reflection) divided by PP: AIx = pressure increase/PP × 100. Because AIx is influenced by heart rate, we estimated AIx adjusted to a heart rate of 75 beats per minute (AIx@HR75).

| Statistical analyses
Data were expressed for continuous variables as mean ± SD and categorical variables as counts and percentages. The Student's t test and Pearson's chi-square test were used to comparing each parameter as needed. The Mann-Whitney U test was used for skewed variables, and Fisher's exact test was used when the expected frequency was lower than 5.
Considering that PWA is affected by physical characteristics, a propensity score matching strategy was used to minimize confounders for adjusted by age, gender, mean blood pressure, heart rate, height, and weight. Participants were matched using 1-to-1 nearest-neighbor matching without replacement. A caliper width of 0.2 of the SD of the logit of the propensity score was used for the developed propensity score.
For all tests, a P value less than .05 was considered statistically significant. All statistical analyses were performed using R statistical

| VVS vs negative HUT test control subjects
We also compared PWA parameters between VVS patients and HUT negative control subjects (Table S2). Although there was no difference F I G U R E 1 Comparison of augmentation index adjusted to a heart rate of 75 beats/minute (AIx@HR75) and pulse wave velocity (PWV) among each group in age between the two groups, we performed matched analysis because gender, height, and weight were significantly different. In a matched analysis, T1, T2, and Tr were comparable between the two groups. However, AIx@HR75 was significantly higher in VVS patients (18.7 ± 13.1% vs 10.2 ± 13.8%, P = .01). In PWV, there was no significant difference (7.0 ± 1.6 m/s vs 6.5 ± 1.3 m/s, P = .14).

T A B L E 2
Baseline characteristics and pulse wave analysis according to the age distribution in the 1:1 matched study population Abbreviations: AIx, augmentation index; AIx@HR75, augmentation index adjusted to a heart rate of 75 beats per minute; AP, augmentation pressure; BMI, body mass index; BP, blood pressure; bpm, beats per minute; BSA, body surface area; ED, ejection duration; MP, mean pressure; PP, pulse pressure; PWV, pulse wave velocity; SEVR, subendocardial viability ratio; T1, time at the first peak/shoulder during systole (outgoing pressure wave); T2, time at the second peak/shoulder during systole (reflected pressure wave); Tr, time to return of the reflected pressure; VVS, vasovagal syncope.

| DISCUSSION
The purpose of this study was to investigate the difference in vascular properties between patients with VVS and healthy subjects using PWA. We observed significant changes in the aortic pressure waveform in patients with VVS. The VVS group had greater aortic stiffness than the control group. Our results indicated that VVS patients have different vascular properties compared to healthy individuals, which supports our hypothesis. Unlike previous studies, our study has novelty in that it showed a difference in vascular properties even after correcting for factors that could affect vascular waveform. [14][15][16] The arterial pressure waveform is determined by the left ventricular stroke volume, the physical properties of the arterial wall, and blood pressure properties. A pressure waveform is initiated when blood exits the heart, and the pressure waveform proceeds faster than the blood flow. The pressure waveform progresses faster as the blood vessel becomes harder and smaller. 13 Arterial stiffness refers to the degree of rigidity caused by the decrease in the elasticity of the arteries. 17 The most important factor in determining arterial stiffness is age. With aging, changes in arterial wall tissues result in decreased elasticity and increased stiffness. 18,19 Arterial stiffness increases with an elevation of blood pressure and other diseases (eg, chronic heart failure, diabetes, and hyperlipidemia) as well as in conditions such as smoking and obesity. 15,20 PWA is a useful tool for noninvasive assessment of central hemodynamics and arterial elasticity indices that analyze the arterial pressure waveform. 8 It is possible to determine important clinical parameters related to vascular stiffness, AIx, and PWV. PWV is the measurement of the speed of the pressure waves that travel along with the arterial segments, indicating the stiffness for a certain distance. On the other hand, AIx is defined as the change in the magnitude of PP caused by the reflected wave, a major measurement of hemodynamics associated with arterial stiffness. Because AIx is influenced by heart rate, the corrected index for heart rate 75 bpm (AI@HR75) is commonly used. 21 Using PWV measurements, a clinician can gauge arterial stiffness that is reflective of the history of the patient's illness and can assess the effect of drug therapy in persons with normal ventricular ejection by measuring aortic AIx. 10,12,13 These F I G U R E 2 The trend of augmentation index adjusted to a heart rate of 75 beats/minute (AIx@HR75) and pulse wave velocity (PWV) according to the age distribution. AIx@HR75 greatly expanded as the age decreases in the vasovagal syncope (VVS) patients. By contrast, PWV was greater with increased age in the VVS patients two parameters are known as independent predictors of cardiovascular disease.
In this study, we found that both PWV and AIx were higher in VVS than in the control group, indicating increasing vascular stiffness.
It is important to emphasize that these modifications were observed at rest in a supine position, without any orthostatic stress. to the site and method of recording pressure waveforms in the present study. 23 We speculated that a decrease in systemic arterial elasticity would be a major factor of VVS at young ages, and impaired compensation of vascular tone due to increased peripheral arterial stiffness would be more dependent on VVS in elderly patients because arterial elasticity necessarily decreases with aging.
In a previous study, peripheral PWA detected a higher stiffness index and a longer time delay between the systolic blood pressure and diastolic blood pressure peak during finger arterial pressure monitoring. 24 Another previously published report describes significant changes in the aortic elastic properties in patients with VVS measured using transthoracic echocardiography. 14 The authors concluded that the aorta is an important modulator of cardiovascular homeostasis.
Their results showed that aortic stiffness index and aortic elastic modulus were higher in patients with VVS compared to healthy individuals. Similarly, using the PWA in our study, we found increased aortic stiffness in VVS patients.
However, merely increased arterial stiffness cannot explain the entire mechanism of VVS because the current opinion suggests that the major pathophysiologic mechanism of VVS is autonomic dysfunction. 4 propensity score matching analysis, and we found that the AIx is consistently increased in VVS patients. Thus, it is further in support of our assertion for pathophysiological relevance. Forth, because of the relatively small number of patients and the single referral tertiary institute data, our study participants may be a skewed and selected population, rather than representing the general population. For this reason, we instituted strict inclusion and exclusion criteria. Our findings would be validated in a larger cohort with multi-center studies.
In conclusion, our study found that patients with VVS have altered aortic pressure waveforms with greater arterial stiffness compared to healthy controls. These findings will help determine the mechanisms involved in the pathophysiology of VVS. Further research is needed to provide more robust information about the direct mechanistic relationship between arterial stiffness and autonomic function in VVS.

ACKNOWLEDGMENTS
This work was supported by Inha University Hospital grant.

CONFLICT OF INTEREST
The authors have no conflict of interest to declare.