Skip to main content
NCBI home page
International Journal of Cardiology. Heart & Vasculature logoLink to International Journal of Cardiology. Heart & Vasculature
. 2026 Feb 10;63:101885. doi: 10.1016/j.ijcha.2026.101885

Comparing Physical, Pharmacological, Pacemaker, and Cardioneuroablation Therapies for Patients with Vasovagal Syncope: A systematic review and network meta-analysis

Haozhe Wang a,1, Yike Zhang a,1, Siyi Liao a,1, Hao Zhang a,1, Xinmiao Huang b,⁎
PMCID: PMC12914290  PMID: 41716718

Abstract

Background

Vasovagal syncope (VVS) represents the most frequent syncope subtype, but evidence comparing treatment strategies remains limited. This network meta-analysis (NMA) evaluates physical, pharmacological, pacemaker, and cardioneuroablation (CNA) therapies for VVS.

Methods

We conducted a Bayesian NMA of randomized controlled trials (RCTs) from PubMed, Embase, and Web of Science. The primary outcome was spontaneous syncope recurrence, and the secondary outcome was head-up tilt test (HUTT) positivity. Network geometry and treatment rankings were evaluated using Surface Under the Cumulative Ranking values.

Results

A total of 49 RCTs involving 2,798 patients assigned to various treatments were included. For spontaneous syncope recurrence, CNA [OR = 0.077, 95% CrI (0.015, 0.403)], pacing [OR = 0.075, 95% CrI (0.023, 0.22)], pharmacological [OR = 0.33, 95% CrI (0.11, 0.94)], and physical [OR = 0.27, 95% CrI (0.13, 0.57)] therapies were all superior to conventional therapy. Regarding HUTT outcomes, only pharmacological [OR = 5.5, 95% CrI (2.6, 12.0)] and physical [OR = 12, 95% CrI (2.9, 50.0)] therapies showed significant differences compared with placebo. Subgroup analysis identified dual-chamber pacing with closed-loop stimulation (DDD-CLS) as the highest-ranked therapy. Midodrine was the superior pharmacological option, and selective serotonin reuptake inhibitors demonstrated efficacy.

Conclusions

This NMA supports a stratified management approach for VVS. Physical and conventional therapies should be first-line. DDD-CLS pacing shows superior efficacy for cardioinhibitory VVS, while midodrine is the preferred pharmacological option. Although CNA demonstrates promise, it is constrained by limited direct evidence and should be considered hypothesis-generating, underscoring the need for head-to-head RCTs with long-term follow-up.

Keywords: Vasovagalsyncope, Networkmeta-analysis, Cardioneuroablation, Drug therapy, Pacemaker

1. Introduction

Transient loss of consciousness events with suspected syncopal characteristics exhibit high prevalence in the general population [1]. Epidemiological studies estimate that approximately half of individuals experience at least one syncopal episode during their lifetime [2]. Reflex syncope constitutes the predominant etiology across all clinical settings and age groups, within this category, vasovagal syncope (VVS) represents the most frequent subtype [3]. VVS typically is characterized by diaphoresis, warmth, nausea, and pallor, associated with vasodepressor hypotension and/or inappropriate bradycardia. The diagnosis is made primarily on the basis of a thorough history, physical examination, and eyewitness observation [2], [4].

Both the American College of Cardiology/American Heart Association/Heart Rhythm Society (ACC/AHA/HRS) and the European Society of Cardiology (ESC) updated their syncope guidelines in 2017 and 2018, respectively [2], [4]. These guidelines strongly advocate for lifestyle modifications in the management of patients with syncope. However, areas such as physical training, pharmacotherapy, and pacemaker therapy remain subjects of ongoing debate and lack full consensus. Since its initial introduction by Pachon et al. in 2005 as a novel therapeutic approach, cardioneuroablation (CNA) has attracted growing attention and has emerged as a potential therapeutic option for VVS [5]. However, evidence of its therapeutic efficacy remains insufficient, and its comparative advantages or disadvantages relative to other treatment modalities require further investigation.

Previous meta-analyses have often been limited to pairwise comparisons of one or two therapeutic categories, lacking a comprehensive ranking of all major treatment classes within a unified framework. One meta-analysis that included pharmacological, pacing, and physical therapies was restricted to within-category pairwise comparisons [6]. They found that selective serotonin reuptake inhibitors (SSRIs), midodrine, and closed-loop stimulation (CLS) pacing showed significant efficacy, with SSRIs outperforming midodrine [6]. Notably, a network meta-analysis focusing exclusively on pharmacological agents concluded that only midodrine demonstrated significant benefits, while fluoxetine appeared beneficial but warranted further investigation [7]. In terms of physical therapy, other studies have confirmed that yoga, physical counterpressure maneuvers, and tilt training can effectively reduce VVS recurrence [8], [9]. However, existing evidence remains conflicting regarding the relative effectiveness of these treatments, and a comparative ranking across all available therapies has been lacking.

Therefore, this study aims to compare the efficacy of various therapies against placebo or conventional therapy, and to mutually compare these therapies using a network meta-analysis approach, in order to better evaluate their relative effects in preventing VVS recurrence. We conducted a systematic review and network meta-analysis of clinical trials assessing the effectiveness of interventions for VVS.

2. Methods

The present study was conducted in strict accordance with the guidance from the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) statement and the PRISMA statement extension for network meta-analysis. The study protocol was prospectively registered on PROSPERO (CRD420251151878) and has been previously published.

2.1. Search strategy and study selection

A comprehensive electronic search was performed across PubMed, Embase, and Web of Science databases until October 6, 2025. The search terms incorporated both Medical Subject Headings (MeSH) and non-MeSH terms related to “Vasovagal Syncope” AND any terms for the following interventions: “Pharmacological”, “Physical”, “Pacemaker”, and “Cardioneuroablation” for the therapies of VVS. The detailed search strategy and corresponding results are provided in Supplementary Table 1.

We included all randomized controlled trials (RCTs) that evaluated at least one therapy compared to placebo, conventional therapy or other therapies in patients diagnosed with VVS based on clinical criteria and/or a positive head-up tilt test (HUTT), and which reported relevant outcomes such as spontaneous syncope recurrence or HUTT results. Studies involving patients with orthostatic hypotension, cardiac syncope, or other specific subtypes of reflex syncope (e.g., situational syncope, carotid sinus syndrome, and non-classical forms of syncope) were excluded. The conventional therapy group providing reassurance about the benign prognosis of VVS, educating patients on common triggers and interventions included advice to increase daily salt and fluid intake, lying down when prodromal symptoms occurred, and/or engaging in physical counterpressure maneuvers (e.g., half-squatting, leg-crossing, calf-raising, and arm-tensing). No restrictions on language were applied.

Two reviewers (HW and YZ) independently screened the titles and abstracts of all identified studies, followed by full-text assessment for eligibility using Rayya (https://www.rayyan.ai/) [10]. In case of any differences, a discussion with a third author (XH) resolved the disagreement. Additional relevant studies were identified by manually reference lists checking.

2.2. Data extraction and risk of bias assessment

After identifying the final set of included RCTs was identified, data extraction was performed duplicated by an initial reviewer (HW) and subsequently verified by two data extractors (HZ and SL). The extraction sheet included (1) first author, publication year, and country of origin; (2) RCT design, sample size in each arm, eligibility criteria, and population characteristics (age, sex, supine blood pressure and heart rate); (3) type of therapies; (4) follow-up duration; and (5) outcome data (number of syncopal events and/or positive HUTTs after initiating therapy) in each arm. The data for the intention-to-treat approach in studies were considered in case of several reporting methods. Missing data were sought directly from study authors.

All risk of bias assessments were duplicated by two reviewers (HW and SL) independently using the Cochrane risk of bias tool. In case of any differences, a discussion with a third author (XH) resolved the disagreement. Quality assessment was performed in the following bias domains: randomization process, deviations from the intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Each domain in addition to the overall quality assessment was classified as “low risk,” “high risk,” or “some concerns.”.

2.3. Statistical analysis

In this study, descriptive statistics were computed to summarize study characteristics. For continuous variables that were reported as medians and interquartile ranges (IQRs) in the original studies, we applied the Method for Unknown Non-Normal Distributions (MLN) to estimate the corresponding mean and standard deviation (SD) [11]. For each dichotomous outcome, the effect size was expressed as an odds ratio (OR) along with its 95% credible interval (CrI). To visualize the available evidence and evaluate the connectivity of the treatment network for each outcome, we generated network diagrams.

We conducted a Bayesian network meta-analysis employing a random-effects model to facilitate simultaneous multi-treatment comparisons in patients with VVS. The analysis was implemented within a generalized linear model framework using a binomial likelihood and a logit link function. All analyses were performed using R statistical software (version 4.4.1; rjags packages), which facilitates Bayesian inference using Markov Chain Monte Carlo (MCMC) sampling. To maintain objectivity and allow the data to dominate the posterior distributions, we specified vague (non-informative) prior distributions for all model parameters. Convergence of the MCMC chains was deemed achieved when trace plots showed good mixing and the Gelman-Rubin statistic was below 1.01 for all parameters. Subsequently, a league table was generated to present all pairwise comparisons between interventions. We also computed the Surface Under the Cumulative Ranking (SUCRA) values and rankograms to identify the optimal therapeutic strategy.

We evaluated consistency by comparing the quantitative estimates from direct and indirect comparisons in accordance with the experimental designs and primary outcomes of the included studies. In addition, node-splitting analysis was performed to assess whether statistical inconsistency was present, with p > 0.05 indicating no significant inconsistency. In this situation, heterogeneity (I2 index statistic) in the study design was used to estimate a data mode for using random-effects models (I2 > 50%). Subgroup analyses were performed based on specific subtypes of interventions, including drug classes and pacing modes. Sensitivity analysis was conducted by comparing results from both random-effects and fixed-effects models. In addition, the leave-on-out method was used to test the sensitivity of the subgroups.

Finally, comparison-adjusted funnel plots were created using Stata (version 17.0) to evaluate potential publication bias.

3. Results

3.1. Study selection, characteristics, and risk of bias assessment

The initial search identified 30,345 records. After the removal of 16,185 duplicates, the titles and abstracts of 14,160 unique records were screened. Subsequently, 104 studies were sought for full-text assessment, of which 8 could not be retrieved. Following a detailed evaluation of the remaining 96 full-text articles, 47 were excluded for various reasons, resulting in 49 studies being included in the final analysis (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Flow diagram of the study selection process for this meta-analysis based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Characteristics of the 49 included studies are summarized in Supplementary Table 1. These studies, published between 1995 and 2025, were conducted primarily in Italy (n = 13), China (n = 7), Canada (n = 6), and the United States (n = 5). Eligibility criteria for the patients (n = 2,798) were based on a history of VVS confirmed by specific tests, mainly HUTT (Supplementary Table 2). Although not a prerequisite for VVS diagnosis, HUTT was utilized in most studies (n = 35). The mean age of all participants was 37.8 ± 13.4 years, and 54.3% were female. Treatment modalities included pharmacological therapy (24 studies), pacemaker implantation (12 studies), and physical interventions (15 studies), while only one study compared CNA with conventional therapy. The recurrence of syncope was measured in 36 studies, and HUTT-induced VVS was assessed in 17 studies.

We acknowledge that the transitivity assumption might be challenged by the fact that CNA is primarily applied to patients with cardioinhibitory VVS, whereas other interventions include mixed populations. No other significant violations of the transitivity assumption were detected across study characteristics (Supplementary Table 2), suggesting comparability of participant features across treatment comparisons. Quality assessment using the Cochrane Risk of Bias Tool indicated that “selection of the reported result” and “measurement of the outcome” had the lowest overall risk of bias, whereas “deviations from intended interventions” yielded the highest proportion of studies with “some concerns.” Unfortunitily, 3 studies was rated as “high risk” in “deviations from intended interventions” due to the open-label design, which introduces potential higher risk for deviations from intended interventions, attrition bias, and missing outcome data. Detailed results of the quality assessment are presented in Fig. 2 and Supplemental Fig. 1.

Fig. 2.

Fig. 2

Open in a new tab

Risk of bias summary for included randomized controlled trials using Cochrane Collaboration tool for assessment of bias in randomized trials.

3.2. Primary network and pair-way meta-analysis

The evidence network for spontaneous syncope recurrence and HUTT positive outcomes is presented in Fig. 3. All interventions were connected, forming a single, coherent evidence network. The number of studies contributing to each direct comparison is indicated on the edges of the network graph.

Fig. 3.

Fig. 3

Open in a new tab

Network plot for spontaneous vasovagal syncope (VVS) recurrence (A) and head-up tilt test (HUTT) positive result (B).

Supplementary Fig. 2 of the network meta-analysis shows that for spontaneous syncope, CNA [OR = 0.077, 95% CrI (0.015, 0.403)], pacing [OR = 0.075, 95% CrI (0.023, 0.22)], pharmacological [OR = 0.33, 95% CrI (0.11, 0.94)], and physical therapies [OR = 0.27, 95% CrI (0.13, 0.57)] were all superior to conventional therapy. In contrast, pharmacological therapy [OR = 4.3, 95% CrI (1.8, 12.0)] and placebo control [OR = 7.6, 95% CrI (3.5, 19.0)] were inferior to pacing. No significant differences were observed among physical therapy, pharmacological therapy, and placebo control. It should be noted that the comparisons for CNA versus pacing and physical therapy versus pacing were each based on a single study, whereas the other results were derived from multiple comparisons. For HUTT outcomes, only pharmacological therapy [OR = 5.5, 95% CrI (2.6, 12.0)] and physical therapy [OR = 12, 95% CrI (2.9, 50.0)] showed significant differences compared with placebo; no significant differences were observed among the other groups (Supplementary Fig. 3). The league table of pairwise comparisons is presented in Supplementary Fig. 4 (A, B).

The cumulative ranking probabilities are detailed in Supplementary Fig. 4(C-F). For spontaneous syncope, CNA demonstrated the highest probability of being the most effective treatment, followed by pacemaker and physical therapy. However, due to a meaningful challenge to the transitivity assumption and may bias indirect comparisons involving CNA, SUCRA values and rankogram may be misleading. Conventional therapy and placebo consistently had the lowest probabilities of being the most effective interventions, indicating their relative inferiority to the active treatments. And for HUTT outcomes, pacemaker and physical therapy both demonstrated the highest probability of being the most effective treatment.

3.3. Convergence and consistency analyses

In this research, convergence was successfully achieved for all model parameters of both outcomes. Trace plots showed good mixing and stationarity after the burn-in period, and density plots indicated unimodal posterior distributions. The Gelman-Rubin statistics for all parameters were below 1.01, confirming convergence. (Supplementary Fig. 5,6).

Moreover, node-splitting models were utilized to assess inconsistencies by testing the differences between direct and indirect effects. For spontaneous syncope, node-splitting analysis identified a significant local inconsistency for the comparison between placebo versus pacemaker (p = 0.03) and placebo versus pharmacological (p = 0.03) (Supplementary Fig. 2). After constructing the node-splitting models for HUTT outcomes, we found that no significant inconsistency existed in this research (p > 0.05) (Supplementary Fig. 3).

3.4. Heterogeneity, subgroup and sensitivity analyses

In the network meta-analysis for heterogeneity assessment, substantial heterogeneity was identified for spontaneous syncope in the comparison between placebo versus pacemaker and placebo versus pharmacological (Supplementary Fig. 2). For the HUTT outcome, significant heterogeneity was confined to the comparisons of physical therapy versus conventional therapy (Supplementary Fig. 3). Accordingly, we performed subgroup analyses and leave-one-out sensitivity analyses for the following pairs: pharmacologic versus placebo and pacing versus placebo (sham pacemaker) in spontaneous syncope, as well as the two heterogeneous pairs in HUTT.

In the subgroup analysis of midodrine, β-blockers, and SSRIs (fluoxetine and paroxetine) versus placebo for spontaneous syncope recurrence, under the random-effects model, both midodrine and SSRIs showed a significant effect [midodrine: OR = 0.41, 95% CI (0.23, 0.74); SSRIs: OR = 0.27, 95% CI (0.13, 0.58)], whereas only β-blockers [OR = 0.78, 95% CI (0.31, 2.00)] did not reach statistical significance (Fig. 4).In the subgroup analysis of different pacemaker modes versus placebo for spontaneous syncope recurrence, conventional pacing modes showed no significant difference [OR = 0.60, 95% CI (0.26, 1.42)], whereas the DDD-CLS (dual-chamber pacing with closed loop stimulation) mode demonstrated a significant advantage [OR = 0.13, 95% CI (0.07, 0.25)] (Fig. 5). Results from the leave-one-out analysis are presented in the corresponding figures (Fig. 4, Fig. 5). In each subgroup, omission of individual studies resulted in a marked reduction in I2 to 0%, although the overall conclusions remained unchanged.

Fig. 4.

Fig. 4

Open in a new tab

Forest plots of subgroup analysis of midodrine, beta-blocks and SSRIs versus placebo for spontaneous syncope recurrence outcome.

Fig. 5.

Fig. 5

Open in a new tab

Forest plots of subgroup analysis of different pacemaker modes versus sham pacemaker for spontaneous syncope recurrence outcome.

For the HUTT outcome, a key study was identified that reduced I2 to 0% in the comparison between physical therapy and conventional therapy (Supplementary Fig. 7). Subgroup analysis of pharmacological therapies showed no statistically significant differences across treatment comparisons (Supplementary Fig. 8).

3.5. Reintegrated network meta-analysis for spontaneous VVS

Based on significant subgroup analysis findings, we separated midodrine, SSRIs and DDD-CLS pacing from their broader intervention categories and refitted the network meta-analysis model for spontaneous vasovagal syncope. The updated network diagram is presented in Supplementary Fig. 9. The revised model demonstrated good convergence, as shown in Supplementary Figs. 10 and 11.

We reintergrated the league table (Fig. 6A) and treatment rankings (Fig. 6 B, C) from the network meta-analysis for spontaneous syncope recurrence. In contrast to the previous analysis, DDD-CLS pacing emerged as the most effective intervention, demonstrating statistically significant superiority compared to physical therapy, other drugs, placebo, and conventional therapy. Midodrine was significantly more effective than placebo, conventional therapy, and other drugs. SSRIs have also demonstrated significant benefits compared to other drugs. The final probability ranking indicated that DDD-CLS had the highest probability of being the optimal intervention, followed by CNA. Midodrine was intermediate, followed by SSRIs and other pacing modes. Physical therapy and other drugs showing the lowest efficacy. It is important to interpret these rankings with caution, particularly for CNA, whose favorable ranking is predominantly driven by indirect evidence and a single RCT. Therefore, greater emphasis should be placed on the effect estimates and their credible intervals (Fig. 6A) rather than on the point estimates of the rankings. The promising position of CNA should be viewed as hypothesis-generating, highlighting the need for direct comparative trials.

Fig. 6.

Fig. 6

Open in a new tab

(A) League table and (B, C) treatment rankings of all comparisons from the network meta-analysis for spontaneous syncope recurrence.

Finally, publication bias was assessed using funnel plots for both the reintegrated network meta-analysis of spontaneous VVS (Supplementary Fig. 12) and the primary network meta-analysis of HUTT outcomes (Supplementary Fig. 13). No substantial asymmetry was observed in either funnel plot, suggesting the absence of obvious publication bias.

4. Discussion

This study incorporated 49 RCTs, encompassing current primary syncope management strategies, including but not limited to supervised or home-based training, pharmacological agents (e.g., β-blockers, α-agonists, SSRIs), pacing modes (e.g., VVI, DDD, DDI, DDD-CLS), and the novel approach of CNA. Utilizing Bayesian methodology, we performed a network meta-analysis to generate pooled estimates and rank therapeutic efficacy across all these therapies. DDD-CLS pacing demonstrated superior efficacy, supporting its use in carefully selected patients with cardioinhibitory VVS. Among pharmacological options, midodrine and SSRIs also demonstrated significant efficacy, whereas β-blockers and fludrocortisone did not show significant benefit. CNA also showed promising results compared to conventional therapy. It is noteworthy that due to the limited number of included RCTs directly comparing CNA with alternative therapies, further investigation is warranted. Additionally, in HUTT-induced syncope, physical and pacing therapy outperformed other therapies.

The ESC 2018 guidelines recommend HUTT for identifying hypotensive susceptibility [4]. The rationale lies in detecting the orthostatic component of VVS, while physical therapies and pacing may offer rapid onset of benefit, consistent with recent meta-analyses focused solely on pacing interventions [9], [12], [13]. Previous meta-analyses have indicated that midodrine and the noradrenaline reuptake inhibitor may be effective in preventing HUTT-induced syncope [14], [15]. Specifically, in subgroup analyses for HUTT, we observed that β-blockers demonstrated a significant therapeutic effect compared to placebo. However, for spontaneous syncope, β-blockers do not demonstrate significant efficacy, which is consistent with the ESC guidelines [4]. For a long time, HUTT outcomes face concerns regarding reproducibility and limited utility in predicting individual therapeutic responses, potentially failing to reliably substitute real-world efficacy [16], [17]. Thus, this study primarily focused on spontaneously occurring syncope events during follow-up.

Physical therapy primarily includes physical counterpressure maneuvers, such as handgrip, tilt training, lower limb compression, as well as supervised or aerobic physical training, the Valsalva maneuver, thigh-length elastic compression stockings (ECS) and yoga. Our study found that physical therapy demonstrated robust and significant efficacy over conventional therapy for spontaneous syncope recurrence, although it did not show a significant difference versus placebo. However, a randomized sham-controlled trial observed that while the overall effect was not significant, the number of VVS episodes was significantly lower in the elastic compression stockings group than in the placebo group during periods when patients actually wore the stockings [18]. Another placebo-controlled study on orthostatic training also found that an enhancement in overall autonomic tone was observed alongside a non-significant trend in symptom improvement [19]. Future large-sample, placebo-controlled trials may be needed to explore the actual effect of physical therapy on spontaneous syncope over long-term follow-up. Along with its marked efficacy in the HUTT outcome, our results, consist with those of previous meta‑analyses on the whole, confirm that physical therapy can reduce the frequency of syncope in patients with recurrent VVS [8], [9]. Because it can effectively abort syncopal episodes, guidelines emphasize that all patients with reflex syncope should first receive the above physical therapy and lifestyle interventions as the cornerstone of first‑line management for VVS [2], [4]. Although the efficacy of physical therapy may not be as pronounced as that of other treatments, it helps reduce syncope recurrence while avoiding the side effects of medications or the risks associated with invasive procedures, and therefore should occupy a primary position in the management strategy for VVS.

In our study, the DDD-CLS pacing mode achieved the highest ranking and demonstrated superiority over traditional pacing modes in subgroup analyses. This finding aligns with the ESC 2018 guidelines and previous meta-analyses [4], [6], [20], [21]. It must be emphasized that DDD-CLS pacing should be strictly reserved for highly selected patients with documented confirmed cardioinhibitory VVS. Due to its characteristic ability to sense sympathetic activation in real-time and provide preemptive intervention, it is particularly suitable for patients with cardioinhibitory VVS who experience spontaneous asystole ≥ 3 s [2], [22]. In contrast, traditional pacing modes operate at a fixed rate and may respond with a delay; their effectiveness in real-world settings requires further investigation [23].

Regarding pharmacological therapy, previous meta-analysis has demonstrated the superiority of midodrine [6], [7], [14], [24]. As an α1-adrenergic receptor agonist, midodrine helps maintain sympathetic tone in peripheral arterioles and veins, increasing venous return and cardiac output, thereby effectively preventing syncope. However, precisely due to its vasoactive properties, the potential risk of supine hypertension limits its clinical application [2]. In the guidelines, fludrocortisone is supported as another main candidate drug with a Class IIb recommendation [4]. However, our analysis did not demonstrate its significant efficacy compared to placebo. It is noteworthy that a consistent result was reported in the network meta-analysis by Behnoush et al., which focused on pharmacological treatments [7]. Despite the failure of the primary analysis in the POST 2 trial to demonstrate unequivocal efficacy, both post-hoc multivariable and on-treatment analyses indicated a statistically significant reduction in syncope likelihood among patients who achieved the target fludrocortisone dose of 0.2 mg daily [25]. Thus, its actual therapeutic effect warrants confirmation by more high-quality studies. Additionally, that same network meta-analysis suggested that for atypical VVS cases accompanied by anxiety or other psychological disorders, fluoxetine might be the best choice [7]. In our study, fluoxetine and paroxetine were combined as SSRIs for a subgroup analysis, which demonstrated significant efficacy. This finding is also consistent with some pairwise meta-analyses [6], [26]. In the subsequent network meta-analysis, the efficacy ranking of SSRIs was relatively low, and significance was only observed when compared with other drug therapies. Further research may be needed to explore their efficacy and applicable populations. In summary, our network meta-analysis supports the guideline recommendations for midodrine, which demonstrated clear efficacy and a high ranking. However, it revealed limited evidence for the efficacy of SSRIs and did not support the significant benefit of fludrocortisone or β-blockers.

As a novel therapeutic approach, CNA achieves permanent cardiac vagal denervation by precisely ablating the epicardial autonomic ganglia, known as ganglionated plexi, of the cardiac autonomic nervous system [5]. Our findings indicate that although it ranks highly, only one study has demonstrated its significant advantage over conventional therapy. Patients in the CNA group exhibited an OR of 0.077 compared to the conventional therapy. However, this finding may be subject to potential overestimation of the OR and warrants further high-quality evidence. However, existing studies have primarily focused on comparing ablation techniques or reporting single-arm investigations of its efficacy and safety, while evidence comparing CNA with other treatment strategies remains limited [27], [28], [29], [30], [31]. To date, only one RCT has compared CNA with optimal non-pharmacological therapy in patients with cardioinhibitory VVS, demonstrating significant therapeutic benefit [32]. Furthermore, the CNA-FWRD registry study is a prospective, observational voluntary registry with a cross-sectional design intended to provide real-world evidence for the long-term efficacy of CNA. However, as a registry study, its evidence strength is lower than that of an RCT and may not fully resolve key questions associated with CNA [33]. Additionally, the CAMPAIGN trial, an international multicenter, prospective, open-label, RCT, is currently underway [34]. The study is expected to provide more robust evidence regarding the comparative efficacy of CNA in the future.

Our findings must be interpreted within the context of important methodological considerations. A key challenge to the transitivity assumption is that the evidence for CNA is derived almost exclusively from trials enrolling patients with documented cardioinhibitory VVS, whereas trials of other therapies included broader VVS populations. This fundamental difference in patient selection may bias indirect comparisons involving CNA and necessitates a cautious interpretation of its relative efficacy. Consequently, the high ranking of CNA, while intriguing, is based on limited and potentially heterogeneous evidence and should be considered exploratory.

When framing our results within the stepped clinical management approach advocated by international guidelines, a clear hierarchy emerges. First-line management for all patients should emphasize physical therapy, which our analysis confirms can effectively abort syncopal episodes and reduce the frequency of recurrent syncope, under the premise of conventional therapy. For patients with recurrent VVS despite active interventions, midodrine represents the pharmacologic agent of choice based on its robust efficacy and high ranking. SSRIs may also offer certain benefits, though further research is required to confirm this. For highly selected patients with refractory, predominantly cardioinhibitory VVS, DDD-CLS pacing should be considered the intervention of choice, given its superior efficacy. In contrast, the promising results for CNA currently position it as an investigational therapy, appropriate primarily within the context of clinical trials until direct comparative evidence becomes available.

4.1. Study limitations

Although this systematic review and network meta-analysis provides a comparison of the efficacy of all available treatments for VVS, several limitations should be acknowledged. First, the evidence for certain interventions, such as CNA, relies on single studies for direct comparisons, introducing a risk of indirect comparison. Second, the inclusion criteria varied across studies, with insufficient control for factors such as frequency of episodes and VVS subtype. Moreover, heterogeneity was difficult to control due to variations in drug dosing, as well as differences in the implementation of physical maneuvers and pacing protocols. Thus, the significant heterogeneity and local inconsistency were identified in several comparisons. Although addressed with random-effects models, subgroup analyses and leave-one-out method, these factors underscore the uncertainty surrounding some point estimates. Third, due to the low recurrence rates (<10%) or even zero events in some studies, ORs were uniformly used for synthesis, which may lead to an overestimation of the effect size compared to risk ratios. This is a particular concern when interpreting the large effect sizes observed for interventions such as CNA and pacing. Finally, most studies lacked long-term follow-up and used HUTT-induced syncope as a surrogate endpoint, which limits the generalizability of the findings to clinical practice. Therefore, future high-quality prospective or RCTs are needed. These should employ strict patient inclusion criteria, compare multiple interventions or include placebo controls, and incorporate long-term follow-up to observe syncope recurrence.

5. Conclusions

In conclusion, this network meta-analysis provides a comparative assessment of all contemporary therapies for VVS. DDD-CLS pacing demonstrated superior efficacy relative to conventional therapy and other active treatments, supporting its use in carefully selected patients with cardioinhibitory VVS. Among pharmacological options, midodrine proved clear efficacy and a high ranking, and SSRIs also demonstrated limited efficacy. CNA also showed promising results compared to conventional therapy. However, its evidence base remains limited, relying predominantly on indirect comparisons and a single RCT, necessitating further validation in head-to-head trials. These findings reinforce a stratified treatment approach: physical and conventional therapies as first-line strategies, followed by midodrine for recurrent cases, with DDD-CLS pacing reserved for highly selected patients within cardioinhibitory VVS. And CNA, as an investigational therapy, appropriate primarily within clinical trial settings at present.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Haozhe Wang: Writing – review & editing, Writing – original draft, Visualization, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Yike Zhang: Writing – review & editing, Validation, Methodology, Investigation, Formal analysis, Data curation. Siyi Liao: Writing – review & editing, Validation, Methodology, Investigation, Data curation. Hao Zhang: Writing – review & editing, Validation, Methodology, Investigation, Data curation. Xinmiao Huang: Writing – review & editing, Validation, Supervision, Resources, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ijcha.2026.101885.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary Data 1
mmc1.docx (18.1MB, docx)

Data availability

Data will be made available on request.

All data analysed in this meta-analysis are derived from previously published studies, which are cited throughout the manuscript. No new data were generated for this study.

References

  • 1.Brignole M., Moya A., de Lange F.J., Deharo J.C., Elliott P.M., Fanciulli A., et al. Practical Instructions for the 2018 ESC guidelines for the diagnosis and management of syncope. Eur. Heart J. 2018;39(21):e43–e80. doi: 10.1093/eurheartj/ehy071. [DOI] [PubMed] [Google Scholar]
  • 2.Shen W.K., Sheldon R.S., Benditt D.G., Cohen M.I., Forman D.E., Goldberger Z.D., et al. 2017 ACC/AHA/HRS guideline for the evaluation and management of patients with syncope: Executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice guidelines and the Heart Rhythm Society. Heart Rhythm. 2017;14(8):e218–e254. doi: 10.1016/j.hrthm.2017.03.005. [DOI] [PubMed] [Google Scholar]
  • 3.Kenny R.A., Bhangu J., King-Kallimanis B.L. Epidemiology of syncope/collapse in younger and older Western patient populations. Prog. Cardiovasc. Dis. 2013;55(4):357–363. doi: 10.1016/j.pcad.2012.11.006. [DOI] [PubMed] [Google Scholar]
  • 4.Brignole M., Moya A., de Lange F.J., Deharo J.C., Elliott P.M., Fanciulli A., et al. 2018 ESC guidelines for the diagnosis and management of syncope. Eur. Heart J. 2018;39(21):1883–1948. doi: 10.1093/eurheartj/ehy037. [DOI] [PubMed] [Google Scholar]
  • 5.Pachon J.C., Pachon E.I., Pachon J.C., Lobo T.J., Pachon M.Z., Vargas R.N., et al. “Cardioneuroablation”–new treatment for neurocardiogenic syncope, functional AV block and sinus dysfunction using catheter RF-ablation. Europace : European Pacing, Arrhythmias, and Cardiac Electrophysiology : Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the European Society of Cardiology. 2005;7(1):1–13. doi: 10.1016/j.eupc.2004.10.003. [DOI] [PubMed] [Google Scholar]
  • 6.Kaza N., Sorbini M., Liu Z., Johal M., Porter B., Nowbar A., et al. Therapeutic options for neurocardiogenic syncope: a meta-analysis of randomised trials with and without blinding. Open Heart. 2024;11(1) doi: 10.1136/openhrt-2024-002669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Behnoush A.H., Yazdani K., Khalaji A., Tavolinejad H., Aminorroaya A., Jalali A., et al. Pharmacologic prevention of recurrent vasovagal syncope: a systematic review and network meta-analysis of randomized controlled trials. Heart Rhythm. 2023;20(3):448–460. doi: 10.1016/j.hrthm.2022.12.010. [DOI] [PubMed] [Google Scholar]
  • 8.Abdelazeem B., Abbas K.S., Manasrah N., Amin M.A., Mohammed S.M., Mostafa M.R. Yoga as a treatment for vasovagal syncope: a systematic review and meta-analysis. Complement. Ther. Clin. Pract. 2022;48 doi: 10.1016/j.ctcp.2022.101579. [DOI] [PubMed] [Google Scholar]
  • 9.Alharbi A., Shah M., Gupta M., Rejent K., Mahmoud M., Alsughayer A., et al. The efficacy of non-pharmacological and non-pacing therapies in preventing vasovagal syncope: Tilt training, physical counter pressure maneuvers, and yoga - a systematic review and meta-analysis. Auton. Neurosci. 2024;251 doi: 10.1016/j.autneu.2023.103144. [DOI] [PubMed] [Google Scholar]
  • 10.Ouzzani M., Hammady H., Fedorowicz Z., Elmagarmid A. Rayyan-a web and mobile app for systematic reviews. Syst. Rev. 2016;5(1):210. doi: 10.1186/s13643-016-0384-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cai S., Zhou J., Pan J. Estimating the sample mean and standard deviation from order statistics and sample size in meta-analysis. Stat. Methods Med. Res. 2021;30(12):2701–2719. doi: 10.1177/09622802211047348. [DOI] [PubMed] [Google Scholar]
  • 12.Brignole M., Deharo J.C., Menozzi C., Moya A., Sutton R., Tomaino M., et al. The benefit of pacemaker therapy in patients with neurally mediated syncope and documented asystole: a meta-analysis of implantable loop recorder studies. Europace : European Pacing, Arrhythmias, and Cardiac Electrophysiology : Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the European Society of Cardiology. 2018;20(8):1362–1366. doi: 10.1093/europace/eux321. [DOI] [PubMed] [Google Scholar]
  • 13.Ruzieh M., Ghahramani M., Nudy M., Naccarelli G.V., Mandrola J., Grubb B.P., et al. The benefit of closed loop stimulation in patients with cardioinhibitory vasovagal syncope confirmed by head-up tilt table testing: a systematic review and meta-analysis. Journal of Interventional Cardiac Electrophysiology : an International Journal of Arrhythmias and Pacing. 2019;55(1):105–113. doi: 10.1007/s10840-019-00531-0. [DOI] [PubMed] [Google Scholar]
  • 14.Lei L.Y., Raj S.R., Sheldon R.S. Midodrine for the prevention of vasovagal syncope: a systematic review and meta-analysis. Europace : European Pacing, Arrhythmias, and Cardiac Electrophysiology : Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the European Society of Cardiology. 2022;24(7):1171–1178. doi: 10.1093/europace/euab323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lei L.Y., Raj S.R., Sheldon R.S. Pharmacological norepinephrine transporter inhibition for the prevention of vasovagal syncope in young and adult subjects: a systematic review and meta-analysis. Heart Rhythm. 2020;17(7):1151–1158. doi: 10.1016/j.hrthm.2020.02.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Moya A., Permanyer-Miralda G., Sagrista-Sauleda J., Carne X., Rius T., Mont L., et al. Limitations of head-up tilt test for evaluating the efficacy of therapeutic interventions in patients with vasovagal syncope: results of a controlled study of etilefrine versus placebo. J. Am. Coll. Cardiol. 1995;25(1):65–69. doi: 10.1016/0735-1097(94)00336-o. [DOI] [PubMed] [Google Scholar]
  • 17.Raviele A., Gasparini G., Di Pede F., Delise P., Bonso A., Piccolo E. Usefulness of head-up tilt test in evaluating patients with syncope of unknown origin and negative electrophysiologic study. Am. J. Cardiol. 1990;65(20):1322–1327. doi: 10.1016/0002-9149(90)91321-v. [DOI] [PubMed] [Google Scholar]
  • 18.Tavolinejad H., Bozorgi A., Emkanjoo Z., Oraii A., Shahabi J., Mollazadeh R., et al. Using Compression Stockings to prevent Recurrence of Vasovagal Syncope: a Randomized Sham-Controlled Trial. J. Am. Coll. Cardiol. 2025;86(6):412–422. doi: 10.1016/j.jacc.2025.05.049. [DOI] [PubMed] [Google Scholar]
  • 19.Tan M.P., Newton J.L., Chadwick T.J., Gray J.C., Nath S., Parry S.W. Home orthostatic training in vasovagal syncope modifies autonomic tone: results of a randomized, placebo-controlled pilot study. Europace : European Pacing, Arrhythmias, and Cardiac Electrophysiology : Journal of the Working Groups on Cardiac Pacing, Arrhythmias, and Cardiac Cellular Electrophysiology of the European Society of Cardiology. 2010;12(2):240–246. doi: 10.1093/europace/eup368. [DOI] [PubMed] [Google Scholar]
  • 20.Varosy P.D., Chen L.Y., Miller A.L., Noseworthy P.A., Slotwiner D.J., Thiruganasambandamoorthy V. Pacing as a Treatment for Reflex-Mediated (Vasovagal, Situational, or Carotid Sinus Hypersensitivity) Syncope: a Systematic Review for the 2017 ACC/AHA/HRS Guideline for the Evaluation and Management of Patients with Syncope: a Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice guidelines and the Heart Rhythm Society. Circulation. 2017;136(5):e123–e135. doi: 10.1161/CIR.0000000000000500. [DOI] [PubMed] [Google Scholar]
  • 21.Kheiri B., Abdelazeem B., Osman M., Dalouk K., Stecker E., Raitt M.H., et al. Meta-analysis of Pacemaker Therapy in patients with Neurocardiogenic Syncope. Am. J. Cardiol. 2021;149:159–160. doi: 10.1016/j.amjcard.2021.03.015. [DOI] [PubMed] [Google Scholar]
  • 22.Russo V, Tomaino M, Parente E, Comune A, Giacopelli D, Napoli P, et al. Temporal relationship between haemodynamic changes and activation of closed-loop stimulation during a tilt-induced vasovagal syncope. Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology. 2024;26(2). [DOI] [PMC free article] [PubMed]
  • 23.Palmisano P, Dell'Era G, Russo V, Zaccaria M, Mangia R, Bortnik M, et al. Effects of closed-loop stimulation vs. DDD pacing on haemodynamic variations and occurrence of syncope induced by head-up tilt test in older patients with refractory cardioinhibitory vasovagal syncope: the Tilt test-Induced REsponse in Closed-loop Stimulation multicentre, prospective, single blind, randomized study. Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology. 2018;20(5):859-66. [DOI] [PubMed]
  • 24.Mhanna M., Beran A., Mhanna A.S., Al-Abdouh A., Jabri A., Numan M.J., et al. Midodrine for Recurrent Vasovagal Syncope: a Systematic Review and Meta-analysis of Randomized Control Trials. Am. J .Ther. 2024;31(1):e87–e91. doi: 10.1097/MJT.0000000000001513. [DOI] [PubMed] [Google Scholar]
  • 25.Sheldon R., Raj S.R., Rose M.S., Morillo C.A., Krahn A.D., Medina E., et al. Fludrocortisone for the Prevention of Vasovagal Syncope: a Randomized, Placebo-Controlled Trial. J. Am. Coll. Cardiol. 2016;68(1):1–9. doi: 10.1016/j.jacc.2016.04.030. [DOI] [PubMed] [Google Scholar]
  • 26.Raj P., Lei L., Flevaris P., Raj S., Sheldon R. Serotonin reuptake inhibition for the prevention of vasovagal syncope: a systematic review and meta-analysis. Clinical Autonomic Research : Official Journal of the Clinical Autonomic Research Society. 2023;33(6):811–819. doi: 10.1007/s10286-023-01000-5. [DOI] [PubMed] [Google Scholar]
  • 27.Aksu T, Brignole M, Calo L, Debruyne P, Di Biase L, Deharo JC, et al. Cardioneuroablation for the treatment of reflex syncope and functional bradyarrhythmias: A Scientific Statement of the European Heart Rhythm Association (EHRA) of the ESC, the Heart Rhythm Society (HRS), the Asia Pacific Heart Rhythm Society (APHRS) and the Latin American Heart Rhythm Society (LAHRS). Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology. 2024;26(8). [DOI] [PMC free article] [PubMed]
  • 28.Armani Prata A., Katsuyama E., Scardini P., Antunes V., Granja J., Coan A.C., et al. Cardioneuroablation in patients with vasovagal syncope: an updated systematic review and meta-analysis. Heart Rhythm. 2025;22(2):526–535. doi: 10.1016/j.hrthm.2024.07.103. [DOI] [PubMed] [Google Scholar]
  • 29.Tu B., Chen A., Cai S., Zhang Z., Zhou L., Lai Z., et al. The Efficacy of Left Atrial vs Biatrial Cardioneuroablation in patients with Vasovagal Syncope: a Randomized Clinical Trial. JACC Clinical Electrophysiology. 2025;11(6):1265–1276. doi: 10.1016/j.jacep.2025.01.019. [DOI] [PubMed] [Google Scholar]
  • 30.Tung R., Pujol-Lopez M., Locke A.H., Alyesh D.M., Sundaram S., Shah A.D., et al. Cardioneural Ablation for Functional Bradycardia and Vasovagal Syncope: Outcomes from the U.S. Multicenter CNA Registry. JACC Clinical Electrophysiology. 2025;11(8):1683–1695. doi: 10.1016/j.jacep.2025.04.012. [DOI] [PubMed] [Google Scholar]
  • 31.Kulakowski P., Baran J., Sikorska A., Krynski T., Niedzwiedz M., Soszynska M., et al. Cardioneuroablation for reflex asystolic syncope: Mid-term safety, efficacy, and patient's acceptance. Heart Rhythm. 2024;21(3):282–291. doi: 10.1016/j.hrthm.2023.11.022. [DOI] [PubMed] [Google Scholar]
  • 32.Piotrowski R., Baran J., Sikorska A., Krynski T., Kulakowski P. Cardioneuroablation for Reflex Syncope: Efficacy and Effects on Autonomic Cardiac Regulation-a Prospective Randomized Trial. JACC Clinical Electrophysiology. 2023;9(1):85–95. doi: 10.1016/j.jacep.2022.08.011. [DOI] [PubMed] [Google Scholar]
  • 33.Aksu T, Tung R, De Potter T, Markman TM, Santangeli P, du Fay de Lavallaz J, et al. Cardioneuroablation for the management of patients with recurrent vasovagal syncope and symptomatic bradyarrhythmias: the CNA-FWRD Registry. Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing. 2025;68(2):183-91. [DOI] [PMC free article] [PubMed]
  • 34.Maimaitijiang P., Tu B., Lai Z., Chen A., Zhang Z., Zhou L., et al. The Efficacy of Cardioneuroablation versus Midodrine in patients with Vasovagal Syncope: Design and Rationale for the CAMPAIGN Trial. Journal of Interventional Cardiac Electrophysiology : an International Journal of Arrhythmias and Pacing. 2025;68(2):257–265. doi: 10.1007/s10840-025-02029-4. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data 1
mmc1.docx (18.1MB, docx)

Data Availability Statement

Data will be made available on request.

All data analysed in this meta-analysis are derived from previously published studies, which are cited throughout the manuscript. No new data were generated for this study.

Articles from International Journal of Cardiology. Heart & Vasculature are provided here courtesy of Elsevier

RESOURCES