Empiric pacing in unexplained syncope and bifascicular block: A systematic review and meta-analysis

Graphical abstract

Abstract

Background

Patients with unexplained syncope and bifascicular block (BFB) are at risk for intermittent high-grade atrioventricular block. Management strategies include empiric pacemaker implantation (EPI) or other alternative approaches (implantable cardiac monitor or electrophysiology study), but the optimal strategy remains uncertain.

Objective

To evaluate the efficacy and safety of EPI compared to alternative approaches in patients with unexplained syncope and BFB.

Methods

We systematically reviewed six studies (2 randomized controlled trials and 4 cohort studies; n = 601) comparing EPI versus alternative strategies in patients with BFB and unexplained syncope. Primary outcomes were predefined as syncope and bradycardia. Secondary outcomes were bradyarrhythmia, cardiovascular death, and device complications. Risk ratios (RR) were synthesized with a random-effects model.

Results

EPI significantly reduced recurrence of syncope (RR 0.48 [0.35, 0.65], p = 3.4E-6), bradycardia (RR 0.06 [0.016, 0.26], p = 7.6E-5), and bradyarrhythmia (RR 0.64 [0.43, 0.96], p = 0.034). Notably, neither device-related complications (RR 1.30 [1.00–1.71], p = 0.06) nor cardiovascular death (RR 0.98 [0.82–1.16], p = 0.82) were significantly increased with EPI.

Conclusion

In patients with unexplained syncope and BFB, EPI significantly reduces recurrent syncope and bradycardia without increasing device-related complications or cardiovascular mortality. These findings support the efficacy and safety of empiric pacing as a reasonable management strategy in this high-risk population.

1. Introduction

A bifascicular block (BFB) is an interventricular conduction abnormality characterized by a conduction impedance in any two of the three main fascicles of the His-Purkinje system: right bundle branch block (RBBB), left anterior fascicle block (LAFB), or left posterior fascicle block (LPFB) [1], [2]. In population studies, the overall prevalence of BFB in adults is estimated at 1–1.5% [3]. Data from China suggests lower rates when broken down by subtype, with RBBB combined with LAFB occurring in about 0.17% of individuals, and combined with LPFB in roughly 0.05% of individuals [4].

Nonetheless, with an overall mortality that ranges from 2 to 14% [3] and syncope affecting more than a third of patients, BFB is far from a benign finding and demands urgent, in-depth investigation [5], [6]. Unexplained syncope in the setting of BFB raises concern for intermittent high-grade atrioventricular block (AVB) as an underlying cause [5]. The presence of BFB on electrocardiogram suggests a significant risk of progression to complete heart block, with some estimates of 40–50% incidence of advanced AVB within two years [7].

Traditional guidelines advocate for a thorough evaluation, e.g. electrophysiology study (EPS) to document advanced conduction abnormality with His-Ventricular interval of 70 ms or greater before committing to pacemaker therapy. According to the European Society of Cardiology (ESC) and the ACC/AHA/HRS guidelines, EPS is a reasonable option (class IIa recommendation, level B evidence) for patients presenting with unexplained syncope and conduction abnormalities. If the EPS is abnormal, pacemaker implantation is indicated (class I recommendation, level B evidence). In some cases, empiric pacing may be considered even without performing an EPS (class IIb recommendation, level B evidence) [1]. Meanwhile, in patients with asymptomatic BFB, implantable cardiac monitoring (ICM) to evaluate for suspected higher-degree AVB is a class IIb recommendation.

Despite these guidelines, the role of empiric pacemaker implantation (EPI) in preventing recurrent syncope among patients with BFB remains a subject of ongoing debate, with limited high-quality evidence to support definitive guidance. Furthermore, the guidelines on this topic are dated, lacking an update since 2018, and do not include robust comparative data². Given the lack of clinical cogency, we conducted a systematic review and meta-analysis comparing EPI with alternative strategies, including EPS and ICM, in patients with unexplained syncope and BFB. Our goal is to evaluate the efficacy of EPI as a reasonable management strategy in this high-risk population without excess harm.

2. Methods

2.1. Information sources

We performed a systematic search in PubMed/MEDLINE, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), and Scopus from database inception to October 1, 2025.

2.2. Search strategy and study selection

We used the search terms “bifascicular block,” “pacemaker,” “implantable cardiac monitor,” “loop recorder,” “electrophysiology study,” “syncope,” including combined queries such as “bifascicular block” AND “syncope.” Two reviewers independently screened titles and abstracts to exclude irrelevant records, such as case reports, reviews, or studies not involving either BFB or syncope. Only studies that included both BFB and syncope were included for quantitative analysis. Disagreements were resolved by discussion and consensus. The remaining articles were retrieved and assessed for eligibility based on prespecified inclusion and exclusion criteria. A PRISMA 2020 flow diagram (Fig. 1) documented records identified, screened, sought for retrieval, assessed for eligibility, as well as reasons for exclusion and studies included.

Fig. 1.

PRISMA 2020 Flow Diagram of Study Selection.

2.3. Data extraction and items

Two reviewers independently extracted study metadata, cohort characteristics (sample size, age, sex), diagnostic criteria for BFB and syncope, intervention and comparator definition (EPI vs. ICM or EPS), follow-up duration, and event counts and totals in each arm for prespecified outcomes: syncope, bradycardia, bradyarrhythmia, cardiovascular death, and device-related complications. Disagreements were resolved by consensus.

2.4. Nomenclature

The data that was extracted from studies was reported heterogeneously, which necessitates proper definitions. BFB was defined on surface ECG as conduction impedance in any two of the three main fascicles of the His-Purkinje system: right bundle branch block (RBBB), left anterior fascicle block (LAFB), or left posterior fascicle block (LPFB). Bradycardia type was not defined explicitly in all studies, but nonetheless in SPRITELY was defined as bradycardia leading to pacemaker implantation [8]. Rivera-López defined bradycardia as significant symptomatic bradycardia [7]. The definition of bradyarrhythmia differed among studies, but as an aggregate involved primarily AVBs and sparsely sinus arrest. Cardiovascular death followed the authors’ adjudication. Device complications included procedural or device-related adverse events reported post-implant.

Because “unexplained syncope” was defined at the individual-study level, we accepted each study’s operational definition. In general, “unexplained syncope” was defined as transient loss of consciousness that remained without an alternative etiology after an initial clinical evaluation as determined by the original investigators. Studies varied in how explicitly they excluded reflex/orthostatic or other non-arrhythmic causes, and the clinical severity of syncope (traumatic versus non-traumatic episodes, presence/absence of prodrome) was not uniformly captured across reports and therefore could not be meta-analyzed as a standalone stratifier. Table S1 shows clinical endpoints and their definitions across studies.

2.5. Assessment of study quality

Randomized trials were appraised with the Cochrane Risk of Bias tool. Observational studies were appraised with the Newcastle-Ottawa Scale. Assessments were performed independently by two reviewers with consensus adjudication.

2.6. Summary measures

The primary effect measure was risk ratio (RR). For each study-outcome combination, we computed the point estimate RR from event counts and totals. The primary analysis pooled all studies together and stratified by outcome.

2.7. Statistical analysis

For all outcomes, we used a random effects model to calculate risk ratio. We assessed between-study heterogeneity using Cochran’s Q test and calculated I² to quantify the proportion of variability due to heterogeneity (Table S2). We inspected the influence of individual studies by leave-one-out checks and by examining consistency of effects across study design (observational vs. randomized control trial). We conducted a sensitivity analysis, thereby reducing heterogeneity and avoiding assumptions. No single study materially altered the direction or significance of the pooled estimates (Table S3).

All processing, statistics, and figure generation were performed in Python 3.13.7 using standard open-source libraries (pandas, numpy, scipy, statsmodels, and matplotlib). All outcome data was analyzed using a random effects model. The Cochran Q test of heterogeneity and the I² of inconsistency were used to assess heterogeneity between studies. Statistically significant heterogeneity was defined as a Cochran Q p-value less than 0.05 or an I² greater than 75%. We did not perform formal statistical testing for publication bias because fewer than 10 studies contributed per outcome, limiting power for such assessments.

3. Results

3.1. Study selection and characteristics

Our search identified 316 reports, of which 24 were evaluated for eligibility (Fig. 1). Of these studies, 6 were excluded because they did not include patients with syncope, 4 were excluded because they did not include patients with BFB, and 7 were excluded because they did not include quantitative outcomes. Six primary studies were analyzed in the final meta-analysis (Table 1) [6], [7], [8], [16], [18], [19]. Two were randomized controlled trials and four were observational cohort or registry studies. The analysis pooled all studies together and stratified by outcome. A protocol for this review was submitted to the International Prospective Register of Systematic Reviews (PROSPERO) (CRD420251162043), and the review was conducted in accordance with PRISMA 2020 guidelines. A PRISMA checklist is provided in the supplementary data.

Table 1.

Characteristics of included studies.

Study	Year	Report design	Follow-up outcomes	ECG method	Follow-up duration (mo.)	Study group, n	Mean age (y)	Female(%)	Bifascicular block, n/N (%)	Mean LVEF (%)
Sheldon et al. (SPRITELY)8	2021	Randomized pragmatic trial	Composite: CV death, syncope, bradycardia requiring PM, device complication	ICM or PM (Empirical PM vs ICM)	Median 33–34	57 PM / 58 ICM	PM 75 ± 9; ICM 78 ± 9	PM 35%; ICM 24%	115/115 (100%)	Preserved (≥35)
Rivera-López et al.7	2020	Prospective cohort	Syncope recurrence, AV block, complications, death	EPS-guided vs Empirical PM	Mean 105 ± 51	36 Empirical PM / 41 EPS- guided	71.7 ± 9.6	39%	77/77 (100%)	≥50
Palmisano et al.6	2022	Prospective multicenter registry	Time to first syncope recurrence	ILR or Pacemaker	Median 33 (IQR 14–52)	123 ILR / 186 PM	77.2 ± 12.2	39.2%	309/309 (100%)	56.6 ± 5.7
Santini et al. (PRESS)16	2013	Randomized single-blind trial	Composite: Syncope, presyncope w/ device intervention, symptomatic AVB	Dual-chamber PM (DDD60 vs DDI30)	24	52 DDD60 / 49 DDI30	77 ± 8	40%	101/101 (100%)	57 ± 10
Doundoulakis et al.18	2023	Prospective observational registry	Syncope recurrence	EPS ± Pacemaker	Mean 46 ± 28	86 PM / 45 No PM	63.7 ± 16.5	33%	37/131 (28.2%)	55 ± 12
Twidale et al.19	NR	Prospective EPS-guided study	Syncope recurrence, mortality	EPS ± Pacemaker	Mean 39 (range 2–125)	45 EPS + Tx / 8 Empiric PM / 40 No Tx	NR	NR	93/93 (100%)	NR

ECG = electrocardiogram; CV = cardiovascular, LVEF = left ventricular ejection fraction; PM = pacemaker; ICM = implantable cardiac monitor; ILR = implantable loop recorder; EPS = electrophysiology study; Tx = treatment; NR = not reported.

3.2. Study quality

We appraised the two randomized controlled trials with the Cochrane Risk of Bias (RoB) tool, and the four cohort studies with the Newcastle-Ottawa Scale (NOS). The overall quality of the studies was very good to excellent. Both randomized trials were judged low risk of bias across all RoB 2.0 domains (randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting), with overall low risk (Table 2). Among cohort studies, each achieved a NOS score of 7/9, reflecting good quality (Table 3).

Table 2.

Risk of bias assessment of randomized controlled trials with the Cochrane risk of bias tool.

	D1	D2	D3	D4	D5	Overall
Sheldon et al. (SPRITELY)8
Santini et al. (PRESS)16

D1: bias arising from the randomization process.

D2: bias due to deviations from intended intervention

D3: bias due to missing outcome data.

D4: bias in measurement of the outcome.

D5: bias in selection of the reported result.

Table 3.

Quality assessment of cohort studies with the Newcastle-Ottawa Scale.

Study		Palmisano et al.6	Rivera-López et al.7	Twidale et al.19	Doundoulakis et al.18
Selection	Representativeness of the exposed cohort	*	*	*	*
	Selection of the nonexposed cohort	*	*	*	*
	Ascertainment of exposure	*	*	*	*
	Demonstration that the outcome of interest was not present at the start of the study
Comparability	Comparability of cohorts based on design or analysis	*	*	*	*
Outcome	Assessment of the outcome	*	*	*	*
	Follow-up duration adequate	*	*	*	*
	Adequacy of follow-up	*	*	*	*
Total Score		7	7	7	7

3.3. Study participants

Across six studies (two randomized controlled trials and four observation studies), we analyzed 601 patients with unexplained syncope and BFB. The two randomized trials restricted enrollment to patients with syncope and BFB and randomized them to EPI versus ICM or programmed minimal backup pacing (DDI30). The cohort studies retrospectively or prospectively followed patients with BFB, with or without EPI, to ascertain prespecified cardiovascular outcomes. The baseline characteristics of the study participants are outlined in Table 1.

Participants were older adults, with the mean age across studies being 74.1 years. Women comprised 36.7% of the study population. Follow-up period varied between studies at the investigators’ discretion. The mean follow-up period of all studies was 41.4 months. Where reported, left ventricular function was preserved. All participants included in this meta-analysis were confirmed to have BFB determined by electrocardiogram. Left bundle branch block (LBBB) was the most prevalent BFB morphology. In the largest included cohort (Palmisano et al. 2022), LBBB accounted for approximately two-thirds of enrolled patients (68.1%), whereas RBBB with LAFB and RBBB with LPFB were less frequent [6]. Other included studies defined BFB by ECG criteria but did not consistently report subtype distributions.

3.4. Primary and secondary outcomes

A total of 6 studies (2 randomized controlled trials and 4 cohort studies) met eligibility criteria and were included in the quantitative analysis. The studies enrolled patients with both BFB and syncope who were managed with EPI or alternative strategies (ICM or EPS). We analyzed three prespecified primary outcomes: recurrent syncope, sinus bradycardia, and progression to bradyarrhythmia. These outcomes were chosen as primary outcomes as they are three of the most common clinical manifestations in patients with syncope and BFB that may necessitate pacemaker implantation. Consequently, these three outcomes were the most commonly reported outcomes among the included studies. Secondary outcomes included cardiovascular death and device complications.

Across all 6 studies, EPI was associated with a 52% pooled reduction in syncope recurrence (RR 0.48 [0.35, 0.65] p = 3.4E-6) when compared to monitoring strategies (Fig. 2). Most individual studies favored EPI, and only two reported no significant difference. There was also a 94% reduction in bradycardia recurrence (RR 0.06 [0.016, 0.26], p = 7.6E-5) (Fig. 3). Furthermore, there was a 36% reduction in progression to bradyarrhythmia (RR 0.64 [0.43, 0.96], p = 0.034) (Fig. 4). Data for cardiovascular death or device complications was limited to two studies. In pooled analyses, there was no statistically significant difference in cardiovascular death (RR 0.98 [0.82–1.16], p = 0.82) or device complications (RR 1.30 [1.00–1.71], p = 0.06) between EPI and alternative monitoring strategies.

Fig. 2.

Random effects model of the likelihood of syncope recurrence in patients with BFB and syncope. Risk ratios are expressed as EPI-to-control, such that values less than 1 favor EPI.

Fig. 3.

Random effects model of the likelihood of bradycardia recurrence in patients with BFB and syncope. Risk ratios are expressed as EPI-to-control, such that values less than 1 favor EPI.

Fig. 4.

Random effects model of the likelihood of bradyarrhythmia recurrence in patients with BFB and syncope. Risk ratios are expressed as EPI-to-control, such that values less than 1 favor EPI.

4. Discussion

To the best of our knowledge, this is the first systematic review and meta-analysis assessing the efficacy and safety of EPI in patients with unexplained syncope and BFB. We found that EPI significantly reduces recurrent syncope, bradycardia, and progression to bradyarrhythmia without a clearly demonstrable increase in device-related complications or cardiovascular mortality.

4.1. Pathophysiology of syncope in bifascicular block

Conduction abnormalities underlying BFB may arise from myocardial ischemia, inflammatory conditions such as myocarditis, or hemodynamic stress from elevated ventricular pressure or volume overload, among other less common causes that impair the His-Purkinje system [9]. Idiopathic progressive fibrosis and sclerosis of the conduction system can be causative, too, irrespective of etiology [10]. The mechanism of progression from BFB to complete heart block is fundamentally similar across all conduction patterns. Whether the block is RBBB with LAFB, RBBB with LPFB, or LBBB, progressive disease of the remaining fascicle due to diffuse degenerative, ischemic, or infiltrative disease can culminate in advanced AVB [11].

The traditional paradigm attributes syncope in BFB primarily to progression to intermittent high-grade AVB [12]. However, Mitro et al. reported that in patients with preexisting conduction disease including BFB, sinus arrest was a more frequent cause of syncope than complete AVB (53% vs 12%, P = 0.03) [13]. The SPRITELY trial further refined this understanding by demonstrating that while bradycardia necessitating pacemaker implantation remained a concern, patients with BFB were also prone to syncope due to a vasovagal response [8]. SPRITELY also showed that although many episodes of syncope were due to complete heart block, smaller fractions of syncope were due to sinus pause, AV asystole, slow atrial fibrillation, and ventricular tachycardia [8]. In fact, SPRITELY found that in 18 patients on ICM with syncope, 15 were found to be arrhythmic, leaving only 16.7% of patients non-arrhythmic. In contrast, Rivera-López et al. found no evidence of ventricular tachycardia in their 41 EPS patients who experienced syncope [7]. Across other studies, an event-level percentage of syncope deemed as non-arrhythmic could not be reliably quantified, as most studies did not report standardized data for syncope mechanism. This brings to surface the core question of whether a specific syncopal episode is truly attributable to intermittent high-degree AVB. While most studies support this traditional paradigm, SPRITELY demonstrates that vasovagal syncope is also a competing source of syncope, one that cannot be managed with pacing. Therefore, empiric pacemaker implantation must be applied with a focused clinical criteria, such that patients demonstrating clear vasovagal physiology are assigned lower priority for empiric pacing.

4.2. Role of EPS and risk-stratified evaluation

Rivera-López et al. found that LBBB was an independent predictor of syncope recurrence (p = 0.033) [7], although no analogous data is available for RBBB subgroups of BFB. In their retrospective observational study, Fumagalli et al. reported a higher prevalence of bradyarrhythmic syncope in older adults (≥75 years), highlighting the multifactorial and age-dependent nature of syncope in this population [5]. Accordingly, management strategies may differ by age, with older patients who have a higher likelihood of degenerative conduction disease favoring empiric pacing, whereas younger patients with BFB and syncope may warrant a more diagnostic-first approach using EPS or prolonged monitoring. In a SPRITELY substudy, Szaszkiewicz et al. also identified asystole, supraventricular tachycardia (SVT), and diabetes as independent correlates of syncope, although SVT and diabetes alone did not predict syncope recurrence [14]. These findings underscore the heterogeneous mechanisms of syncope in BFB and the challenges of risk stratification in this population.

Importantly, EPS may remain clinically valuable in select patient subsets not uniformly represented in the included studies. Patients with markedly widened bundle branch block, syncope occurring without prodromal symptoms, or heart failure with mildly reduced ejection fraction who do not otherwise meet criteria for implantable cardioverter-defibrillator therapy may benefit from EPS to exclude ventricular tachyarrhythmias, including bundle branch re-entrant ventricular tachycardia. In SPRITELY, patients with LVEF 35–50% were specifically recommended to undergo EPS to rule out monomorphic VT rather than to screen for infra-Hisian disease, reflecting the need to exclude malignant ventricular mechanisms in intermediate risk subgroups even when the overarching question is bradyarrhythmic syncope. In these situations, a diagnostic strategy incorporating EPS to rule out ventricular arrhythmia, followed by pacemaker implantation when appropriate, could be an individualized decision rather than a competing alternative to empiric pacing.

Accordingly, our findings should not be interpreted as negating the role of EPS, but rather as contextualizing its utility within a larger risk-stratified framework. While empiric pacing appears effective in reducing recurrent syncope and bradyarrhythmic events at a population level, the current evidence base, composed largely of observational findings with limited randomized confirmation, should be viewed as hypothesis-generating rather than paradigm-shifting. These results support empiric pacing as a reasonable management option in carefully selected high-risk patients, while strengthening the need for prospective trials designed to better define subgroups in whom EPS-guided evaluation or alternative strategies may be superior.

4.3. Comparative evidence supporting empiric pacing

Current guideline recommendations for pacemaker implantation emphasize documentation of high-grade AVB on surface electrocardiogram or EPS, either during follow-up or at symptom recurrence, rather than systematic reliance on long-term cardiac monitoring [11], [15]. Rivera-López et al. demonstrated in a prospective cohort study that EPI was associated with significantly reduced syncope recurrence (5.6% vs 29.3%, p = 0.02) compared to EPS-guided management, with similar mortality and complication rates. Notably, 55% of patients with negative EPS subsequently experienced recurrent syncope [7]. The most recent meta-analysis evaluating EPS in this context reported a negative predictive value of only 71% for progression to AVB [15], reinforcing the limitations of EPS as a sole management strategy for pacemaker implantation decisions.

No robust evidence currently supports medical therapy to reverse BFB once structural conduction disease is established [15]. After ruling out reversible causes of fascicular blocks, the two major options that physicians face are EPI or prolonged monitoring (e.g., ICM). Palmisano et al. found that the major determinants impacting the decision for empiric pacemaker implantation were prolongation of the PR interval and higher comorbidity status (age and frailty) [6], [17]. Notably, despite exhibiting more severe baseline comorbidities, patients who underwent EPI experienced fewer episodes of traumatic syncope. These findings align with a risk-adapted approach in which empiric pacing is favored for patients with higher pretest probability of bradyarrhythmic syncope.

4.4. Evidence from randomized trials

The PRESS trial reinforced this evidence, showing that patients with BFB and unexplained syncope randomized to active empiric dual-chamber pacing (DDD60) had a 68% reduction in a composite endpoint of syncope/presyncope, presyncope with device intervention, or symptomatic AVB, compared to minimal backup pacing (DDI30) [16]. At 2 years, the composite endpoint occurred in 13.5% of actively paced DDD60 patients versus 32.6% in minimally paced DDI30 patients (p = 0.042), underscoring the superiority of therapeutic pacing and the value of EPI [16]. Our pooled analysis corroborates a clinically meaningful reduction in syncope rates, irrespective of whether episodes were ultimately bradyarrhythmic or reflex in mechanism. A significant 94% decrease in bradycardia events was also observed in our analysis. Most of the included studies included older adults, where non-arrhythmogenic causes of syncope may have acted as a confounding factor. Even so, the majority of studies reported episodes of bradycardia that ultimately required pacemaker implantation.

The SPRITELY trial used a composite endpoint of cardiovascular death, syncope, bradycardia requiring pacemaker, and device complications [8]. In this trial, there was significant crossover from the ICM to pacemaker arm due to symptom development, and all patients in the non-intervention arm ultimately received pacemakers, raising the question of whether empiric pacing at the outset may be more appropriate. It is unclear whether deferring pacing meaningfully reduces device exposure or merely delays necessary therapy. The inclusion of patients with suspected vasovagal syncope in SPRITELY introduces confounding, whereas the PRESS trial mitigated this by excluding such patients with tilt-table testing [16]. PRESS also employed a control group of DDI30 for safety, which allowed minimal pacing and may partly attenuate observed treatment differences. Furthermore, PRESS is the only study that evaluated presyncope, demonstrating that EPI reduced presyncope rates, although it had no effect on syncope itself. Nonetheless, EPI led to decreased rates of a primary endpoint of presyncope/syncope (hazards ratio 0.43 [0.25–0.78], p = 0.0053) [16]. Collectively, this data supports the clinical benefit and practicality of empiric pacing in this high-risk population.

4.5. Implications for clinical practice, policy, and future research

These findings support empiric pacemaker implantation as a reasonable management option for selected patients with unexplained syncope and BFB. From a policy perspective, the pacing benefit observed may warrant possible reassessment of guideline recommendations that more clearly define patient subgroups most likely to benefit from early pacing. Future research should focus on prospective trials that compare empiric pacing with EPS or ILR-guided strategies and their effects on clinically meaningful outcomes such as syncope, bradycardia, and progression to bradyarrhythmia.

5. Limitations

The predominance of nonrandomized evidence introduces residual confounding and selection bias that cannot be fully mitigated with aggregate data. For the syncope endpoint, the two RCTs (SPRITELY and PRESS) show a null association for EPI and syncope recurrence. However, observational studies overwhelmingly support EPI’s role in reduced syncope recurrence such that the pooled data demonstrates a significant reduction in events (Fig. 2). The significant pooled reduction in syncope is therefore partly driven by observational evidence and should be interpreted cautiously.

Furthermore, definitions of syncope, bradycardia, and bradyarrhythmia, as well as follow-up durations and endpoint outcomes varied across studies, which to some extent may contribute to clinical and methodological heterogeneity. For our three primary outcomes, calculated I² shows negligible heterogeneity for bradycardia (0%), low-moderate heterogeneity for syncope (43.20%), and moderate heterogeneity for bradyarrhythmia (53.24%) (Table S2), reflecting some variability but no major inconsistency. With few contributing studies per outcome, Cochran’s Q also has limited power and I² estimates may be imprecise.

Empiric pacing's principal drawback is device-related harm, such that procedural complications and longer-term risks must be weighed against any reduction in syncope and bradycardia events. Only two studies reported cardiovascular mortality and device complications, yielding sparse evidence [7], [8]. Of the 601 patients in the pooled data, only 192 were analyzed for cardiovascular mortality and device complications. In individual studies, SPRITELY observed a higher rate of device-related complications in the pacemaker arm, but with imprecise, wide confidence intervals. By contrast, Rivera-López reported no significant difference in device complications between groups [7]. Consequently, the pooled estimates for these endpoints are underpowered and should be interpreted cautiously. Although the pooled confidence interval for device complications includes the null value with statistical insignificance (RR 1.45 [0.67, 3.11], p = 0.342), the lack of power precludes a definitive conclusion of fewer device complications with empiric pacing.

Furthermore, outcomes within these studies may not be equivalently weighted in terms of clinical importance. Recurrent syncope without injury is not comparable to serious device-related adverse events such as pericardial tamponade requiring invasive management. Regardless, SPRITELY has demonstrated that there was substantial crossover from the ICM arm to pacemaker due to disease progress, and all patients in the non-intervention arm (100%) ultimately received pacemakers [8]. Similarly, in Palmisano et. al, 35 of 49 patients (71.4%) receiving an ILR eventually required pacemaker implantation [6]. Rivera-López et al. found that 12 of 18 patients (66.7%) with an ILR required pacemaker implantation on follow-up [7]. These results show that even patients managed by monitoring were eventually paced and exposed to pacemaker-related procedural risks, prompting the broader question of whether earlier pacemaker implantation would have been the more appropriate initial strategy.

6. Conclusion

This systematic review and meta-analysis demonstrates that EPI in patients with unexplained syncope and BFB is associated with substantial reduction in recurrent syncope and bradycardia, without definite evidence of increased device-related complications or cardiovascular mortality. Deferring intervention in favor of prolonged monitoring may expose high-risk patients to recurrent syncope, bradycardia, and bradyarrhythmia, with the potential for serious life-threatening events. By synthesizing the available evidence, our findings support early empiric pacing as a viable strategy in high-risk BFB patients experiencing syncope, laying the groundwork for further prospective clinical trials.

CRediT authorship contribution statement

Md Fahim: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Daoud Eldawud: Writing – review & editing, Writing – original draft, Validation, Resources, Methodology, Investigation, Formal analysis, Data curation. Nischal Sharma: Writing – review & editing, Writing – original draft, Data curation. Ahmad Al-Abdou: Writing – review & editing. Ahmad Alkhatib: Writing – original draft. Mohammad Abdallah Omar: Writing – review & editing. Roy Wang: Writing – review & editing. Asher Gorantla: Writing – review & editing. Peter Khouri: Writing – review & editing, Data curation. Ahmad Jallad: Writing – review & editing, Supervision. Adam S. Budzikowski: Writing – review & editing, Supervision.

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.

Appendix A. Supplementary data

The following are the Supplementary data to this article: