Contemporary Imaging Modalities for Transcatheter Interatrial Communication Closure: A Comparative Meta‐Analysis of Intracardiac and Transesophageal Echocardiography

ABSTRACT

Background

Intracardiac echocardiography (ICE) is increasingly used to guide cardiovascular interventions. This study compared the performance of ICE with transesophageal echocardiography (TEE) in device closures of interatrial communications.

Methods

A systematic search through PubMed, Scopus, Cochrane Library, and Europe PMC identified clinical studies directly comparing ICE and TEE as imaging guidance modalities for percutaneous closure of ASD or PFO in pediatric and adult populations. The key outcomes encompassed procedural time, fluoroscopy time, procedural success, complete closure at the latest follow‐up, and major and minor complications. Pooled mean differences (MD) and risk ratios (RR) with corresponding 95% confidence intervals (CI) were measured by applying the random‐effects model.

Results

Twelve studies comprising 5054 participants (ICE = 2480; TEE = 2574) were included in the quantitative analysis. Meta‐analysis demonstrated a numerically shorter procedural time (MD −20.80 min; 95% CI −29.84 to −11.75; p < 0.00001), but not fluoroscopy time (MD −1.17 min; 95% CI −2.64 to 0.30; p = 0.12), in the ICE group compared with the TEE group. Procedural success rates and complete closure at the latest follow‐up were comparable between both imaging strategies. In safety analyses, ICE was associated with a meaningful reduction of risk of major complications compared to TEE (RR 0.77; 95% CI 0.64 to 0.92; p = 0.004). Comparable findings were observed in the incidence of minor complications.

Conclusion

ICE represents a promising alternative imaging modality for guiding transcatheter closure of ASD and PFO. Compared with TEE, ICE is associated with a numerically reduced procedural duration and a lower incidence of major complications, while maintaining comparable procedural success and long‐term closure rates.

Intracardiac echocardiography (ICE) is an effective alternative to transesophageal echocardiography (TEE) for transcatheter ASD/PFO closure. In 12 studies involving 5054 patients, ICE reduced procedural time, fluoroscopy exposure, and major complications while maintaining comparable procedural success and complete closure rates, supporting improved procedural efficiency and safety.

Abbreviations

AF

atrial fibrillation

AKI

acute kidney injury

ASD

atrial septal defect

ICE

intracardiac echocardiography

PFO

patent foramen ovale

PPM

permanent pacemaker

TEE

transesophageal echocardiography

TIA

transient ischemic attack

1. Background

Atrial septal defect (ASD) and patent foramen ovale (PFO) are among the most prevalent subtypes of interatrial communications. ASD constitutes a true structural deficiency of the atrial septum, most commonly involving the ostium secundum region, and may result in chronic right‐sided volume overload, and eventual heart failure if left untreated [1]. In contrast, PFO represents a persistent, tunnel‐like interatrial communication arising from incomplete postnatal fusion of the septum primum and septum secundum. The global prevalence of ASD has been estimated at approximately 0.88 per 1000 live births, whereas PFO is observed in 15%–35% of the general population [2, 3, 4]. Although individuals with PFO are typically asymptomatic, the condition has been linked with a heightened risk of cryptogenic stroke and other paradoxical embolic events. Conversely, ASD more frequently leads to progressive hemodynamic sequelae over time, reflecting its nature as a true anatomical defect rather than a potential shunt.

Intraprocedural echocardiographic imaging is integral to transcatheter closure of interatrial communications, enabling accurate assessment of defect size, evaluation of septal rim adequacy, real‐time guidance of device deployment, and prompt identification of procedural complications. Transesophageal echocardiography (TEE) has historically been regarded as the reference standard for imaging guidance due to its superior spatial resolution and detailed visualization of interatrial anatomy [5]. However, TEE frequently necessitates general anesthesia and the presence of a dedicated echocardiographer, potentially increasing procedural complexity. Intracardiac echocardiography (ICE) has emerged as a viable alternative imaging modality, offering high‐resolution, real‐time visualization from within the cardiac chambers. Its use avoids the needs for general anesthesia and may streamline workflow in the catheterization laboratory [6, 7, 8]. Contemporary evidence indicates that ICE‐guided closure demonstrates a safety and efficacy profile comparable to that of TEE‐guided interventions, supporting its role as an effective imaging strategy in appropriately selected patients [9, 10, 11].

In light of the increasing adoption of ICE and the heterogeneity in imaging guidance strategies, this meta‐analysis was conducted to systematically compare ICE‐ and TEE‐guided transcatheter closure of ASD and PFO, with particular emphasis on procedural outcomes. The primary endpoints include procedural success, total procedural duration, fluoroscopy time, periprocedural complications, and the requirement for general anesthesia. The objective is to determine whether ICE confers advantages in procedural efficiency and safety over TEE within contemporary clinical practice.

2. Methods

2.1. Study Protocol

The study protocol for this review4523169623 has been prospectively registered in the PROSPERO database with the registration number of CRD420251150110. The systematic review and meta‐analysis were carried out according to the updated Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) 2020 statement [12].

2.2. Search Strategy

A comprehensive search was independently executed by three investigators (DWA, BBB, and ASP) across multiple search engines, namely PubMed, Scopus, Cochrane Library, and Europe PMC, to obtain eligible studies comparing ICE and TEE guidance in percutaneous closure of ASD and PFO. The search was restricted to articles published in English from database inception through January 2026 to minimize potential language bias. A predefined Boolean search strategy was developed using combinations of relevant free‐text terms as follows: ((intracardiac echocardiography) OR (intracardiac) OR (ICE) OR (echocardiographic guidance) OR (echocardiography guidance)) AND ((transcatheter closure) OR (percutaneous closure) OR (device closure) OR (device closures) OR (atrial septal defect closure) OR (patent foramen ovale closure) OR (closure) OR (closures))) AND ((atrial septal defect) OR (atrial septal defects) OR (ASD) OR (patent foramen ovale) OR (PFO) OR (interatrial communication) OR (interatrial communications)).

2.3. Eligibility Criteria

Prespecified eligibility criteria were determined a priori before commencement of the literature search. Studies were judged to be eligible if they met the inclusion criteria as follows: (1) evaluated the comparative effectiveness of ICE versus TEE for guidance during transcatheter closure of ASD or PFO; (2) enrolled either pediatric or adult populations; (3) adopted any study design, namely randomized controlled trials (RCT) or observational studies; and (4) reported at least one of the predefined study outcomes, namely: (a) procedural time, (b) fluoroscopy time, (c) procedural success rate, (d) complete closure at the latest follow‐up, (e) major complications, and (f) minor complications. Major complications were defined as cardiac erosion, pericardial effusion or cardiac tamponade, device embolization or migration, vascular complications, stroke or transient ischemic attack (TIA), new‐onset atrial fibrillation (AF), permanent pacemaker (PPM) implantation, acute kidney injury (AKI), or major bleeding requiring blood transfusion. Minor complications included access‐site hematoma, transient arrhythmias, and migraine. The exclusion criteria encompassed: (1) review articles, case reports, editorials, brief communications, letters, book chapters, or clinical guidelines; (2) studies evaluating cardiovascular interventions other than ASD or PFO device closure; and (3) non‐English publications.

2.4. Study Selection and Data Extraction

All identified records from the literature search were collected in the Rayyan platform for subsequent screening process. Duplicate entries were removed prior to screening. Subsequently, four independent investigators (DWA, AD, BBB, ASP) screened the relevancy of the titles and abstracts. Full‐text articles of the remaining studies were then retrieved and evaluated for compliance with the eligibility criteria. Essential data from the included studies were then extracted into a structured data collection form using Google Sheets. Extracted variables comprised study ID, study center, number of participants for each intervention arm, mean age, sex distribution, number of ASD or PFO cases, defect size as assessed by TEE, duration of follow‐up, and results for each prespecified outcome. Continuous variables were reported as mean values with the corresponding standard deviations (SDs). When studies reported medians and interquartile ranges (IQRs), the median was considered an estimate of the mean, and the IQR was converted to an SD by dividing by 1.35. For studies evaluating the same intervention across multiple groups, data were pooled using the methods described elsewhere [13]. In cases of incomplete or missing data, corresponding authors were contacted for clarification; studies without author response were excluded from the quantitative synthesis. Any disagreements were discussed until a consensus was achieved.

2.5. Risk of Bias Assessment

The risk of bias for all included studies was evaluated by three independent investigators (RKM, BBB, and BYP) according to study design. RCTs were appraised using the Cochrane Risk of Bias Tool 2 (RoB 2), which examines potential bias across six methodological domains. Observational studies were appraised by employing the Risk Of Bias In Non‐Randomized Studies—of Interventions (ROBINS‐I), which assesses bias across eight domains. The plots for risk of bias evaluation were generated through robvis tool. The detailed criteria and interpretation framework for both RoB 2 and ROBINS‐I have been described previously in their respective methodological guidance documents [14, 15].

2.6. Statistical Analysis

Statistical analyses were performed by employing Review Manager (RevMan version 5.4.1) and Stata (version 18; StataCorp, College Station, TX, USA). All pooled estimates were measured by applying random‐effects model, irrespective of the degree of statistical heterogeneity. Continuous and dichotomous outcomes were assessed using pooled mean differences (MDs) and risk ratios (RRs), respectively, with corresponding 95% confidence intervals (CIs). The outputs of the meta‐analyses are visually displayed using forest plots. Statistical significance was confirmed with a p‐value of < 0.05. Statistical heterogeneity was quantified using the I² statistic, with a threshold of > 50% indicating relevant heterogeneity. To investigate the potential sources of heterogeneity, a subgroup analysis restricted to studies involving exclusively patients with ASD was performed across all study outcomes. Publication bias was evaluated through Egger's and Begg's rank correlation test, with a p‐value ≥ 0.05 suggesting no relevant publication bias. Sensitivity analyses were carried out for each outcome using a leave‐one‐out method to evaluate the robustness of the statistical model. The model was judged to be robust when statistical significance and the direction of the results were consistent throughout sequential exclusion of individual studies.

3. Results

3.1. Literature Search and Study Selection

A total of 1676 records were identified through the predefined database search. Following the removal of 396 duplicate entries, 1280 records were screened for the relevance of the title and abstract. Of these, 1260 were excluded due to irrelevant titles or abstracts. Twenty full‐text articles were retrieved and evaluated based on the eligibility criteria. After full‐text evaluation, 8 studies were excluded for reasons specified in the PRISMA flow diagram. Ultimately, 12 studies were included in the meta‐analysis. The overall study selection process is summarized in Figure 1.

FIGURE 1.

Study selection flow illustrated by PRISMA flow diagram.

3.2. Baseline Characteristics of Included Studies

A total of 12 studies comprising 5054 participants (ICE = 2480; TEE = 2574) were included in the quantitative analysis [9, 10, 11, 16, 17, 18, 19, 20, 21, 22, 23, 24]. Of these, 11 were observational studies and 1 was a RCT. One observational study employed propensity score matching (PSM), resulting in two cohorts with balanced baseline characteristics. Geographically, two studies were performed in the United States, two in Italy, and one each in Germany, France, China, and Taiwan, while two studies originated from Japan and two from South Korea. Sample sizes ranged from 42 to 3308 participants per study. Both pediatric and adult populations were represented, with mean ages ranging from 15.3 to 56 years. Regarding clinical indications, seven studies exclusively enrolled patients with ASD, two focused solely on PFO, and three included both ASD and PFO populations. The follow‐up duration varied between six and thirty‐nine months. The baseline characteristics of the included studies are summarized in Table 1.

TABLE 1.

Baseline characteristics of included studies.

Study ID	Study Design	Study Center	Intervention Arm	Number of Participants	Mean Age (years) (mean ± SD)	Female (n, %)	Assessed Pathology (n, %)	Defect Size (mm) (mean ± SD)	Follow up Duration (months) (mean ± SD)
Alqahtani [16]	Observational, PSM	USA	ICE	1659	53 ± 16	711 (42.9)	NR	NR	NR
			TEE	1659	53 ± 16	710 (42.8)
Bartel [17]	RCT	Germany	ICE	50	46 ± 13	36 (45)	ASD = 12 (15%) PFO = 68 (85%)	NR	6
			TEE	30
Boccalandro [18]	Observational, prospective	USA	ICE	21	50 ± 16	12 (57.14)	ASD = 5 (11.9%) PFO = 37 (88.1%)	NR	6
			TEE	21	48 ± 21	11 (52.38)
Chien [19]	Observational, retrospective	Taiwan	ICE	128	20.9 ± 16.8	NR	ASD = 292 (100%)	14.0 ± 5.9	NR
			TEE	142	15.3 ± 14.9			12.9 ± 5.6
De Cillis [20]	Observational, retrospective	Italy	ICE	57	43 ± 15	43 (75)	ASD = 90 (100%)	NR	6
			TEE	33	42 ± 17	23 (70)
Kim [21]	Observational, retrospective	South Korea	ICE	237	24.2 ± 19.6	172 (72.57)	ASD = 560 (100%)	NR	NR
			TEE	323	28.4 ± 21.5	111 (34.37)
Moon [11]	Observational, retrospective	South Korea	ICE	25	57 ± 7	7 (28)	PFO = 74 (100%)	NR	NR
			TEE	49	47 ± 10	16 (33)
Reibel [22]	Observational, prospective	France	ICE	46	47.8 ± 10.8	22 (47.8)	PFO = 162 (100%)	NR	6
			TEE	116	48.7 ± 13.6	45 (38.8)
Shimizu [23]	Observational, retrospective	Japan	ICE	51	53 ± 15.75	29 (57)	ASD = 92 (100%)	15.7 ± 4.43	12
			TEE	41	47 ± 19.5	26 (63)		13.4 ± 5.13
Wang [24]	Observational, retrospective	China	ICE	23	46.9 ± 18.2	11 (47.8)	ASD = 46 (100%)	NR	NR
			TEE	23	41.8 ± 18.7	11 (47.8)
Yamano [9]	Observational	Japan	ICE	53	47 ± 20	NR	ASD = 152 (100%)	12 ± 1.25	39 ± 31.1
			TEE	98	56 ± 21.5			18.3 ± 2.56
Zanchetta [10]	Observational	Italy	ICE	44	42.56 ± 17.78	32 (72.73)	ASD = 168 (100%)	NR	12
			TEE	125	40.5 ± 20.6	91 (73.39)

Abbreviations: ASD, atrial septal defect; ICE, intracardiac echocardiography; PFO, patent foramen ovale; PSM, propensity score‐matching; TEE, transesophageal echocardiography.

3.3. Risk of Bias Assessment

Based on the risk‐of‐bias assessment, the majority of the included studies were deemed low risk. Nevertheless, most observational studies did not explicitly report blinding of outcome assessors to the allocated intervention, which may introduce detection bias. In addition, the included randomized controlled trial did not provide sufficient details regarding the randomization procedure or the blinding of study personnel, resulting in some concerns related to methodological rigor. As a result, these accumulative findings might pose the statistical model to increased risk of bias. A comprehensive summary of the risk‐of‐bias evaluation is presented in Figure 2.

FIGURE 2.

Risk of bias assessment using RoB2 and ROBINS‐I tools.

3.4. Meta‐analysis

Meta‐analysis suggested that ICE may be associated with a shorter procedural time than TEE (MD −20.80; 95% CI −29.84 to −11.75; I ² = 99%; p < 0.00001; Figure 3). In contrast, fluoroscopy time was insignificantly reduced in the ICE group (MD −1.17; 95% CI −2.64 to 0.30; I ² = 97%; p = 0.12; Figure 3). The I² values indicated substantial heterogeneity for both procedural endpoints despite the statistical significance for the procedural time. In contrast, ICE and TEE showed comparable rates of procedural success (RR 1.00; 95% CI 0.99 to 1.01; I ² = 0%; p = 0.80; Figure 4) and complete closure at the latest follow‐up (RR 1.00; 95% CI 0.98 to 1.01; I ² = 1%; p = 0.66; Figure 4), with no significant heterogeneity observed for either outcome. With respect to safety, the ICE group was linked with a statistically lower incidence of major complications compared with the TEE group (RR 0.77; 95% CI 0.64 to 0.92; I ² = 0%; p = 0.004; Figure 5). However, the incidence of minor complications did not differ significantly between groups (RR 0.73; 95% CI 0.43 to 1.25; I ² = 0%; p = 0.25; Figure 5). No relevant heterogeneity was detected across the safety outcomes. The summary of study findings is provided in Table 2.

FIGURE 3.

Forest plots for procedural outcomes. Mean differences were calculated using the random‐effects model. (a) Procedural time. (b) Fluoroscopy time. ICE, intracardiac echocardiography; TEE, transesophageal echocardiography.

FIGURE 4.

Forest plots for efficacy outcomes. Risk ratios were calculated using the random‐effects model. (a) Success rate. (b) Complete closure at the Latest Follow‐Up. ICE, intracardiac echocardiography; TEE, transesophageal echocardiography.

FIGURE 5.

Forest plots for safety outcomes. Risk ratios were calculated using the random‐effects model. (a) Major complications. (b) Minor complications. ICE, intracardiac echocardiography; TEE, transesophageal echocardiography.

TABLE 2.

Summary of study findings across all outcomes.

Outcomes	Number of Studies	ICE (n)		TEE (n)		Model	MD	95% CI	p‐Value of Heterogeneity	p‐Value of Egger's Test	p‐Value of Begg's Test	p‐Value
Procedural time	10	775		799		Random	−20.80	−29.84 to −11.75	< 0.00001	0.6025	0.8580	< 0.00001
Fluoroscopy time	10	768		816		Random	−1.17	−2.64 to 0.30	< 0.00001	0.4617	0.5915	0.12
Outcomes	Number of Studies	ICE		TEE		Model	RR	95% CI	p‐Value of Heterogeneity	p‐Value of Egger's Test	p‐Value of Begg's Test	p‐Value
		Event, n	Total, n	Event, n	Total, n
Success rate	7	663	670	705	714	Random	1.00	0.99 to 1.01	0.99	0.8140	0.5480	0.80
Complete closure at the latest Follow‐Up	7	543	558	569	592	Random	1.00	0.98 to 1.01	0.42	0.3802	1.0000	0.66
Major complications	11	180	2457	242	2551	Random	0.77	0.64 to 0.92	0.96	0.4965	0.2129	0.004
Minor complications	7	26	628	30	546	Random	0.73	0.43 to 1.25	0.65	0.8365	1.0000	0.25

Abbreviations: ICE, intracardiac echocardiography; TEE, transesophageal echocardiography.

3.5. Subgroup Analysis

A series of subgroup analyses was performed in studies exclusively involving ASD populations. Overall, the subgroup analyses yielded findings that were largely consistent with the primary analysis, with the exception of the major complication outcome. Procedural time remained significantly shorter in the ICE group than in the TEE group, although substantial heterogeneity persisted (MD −18.50; 95% CI −29.66 to −7.34; I ² = 99%; p = 0.001; Figure S1). Likewise, the findings for fluoroscopy time were consistent with those of the overall analysis, demonstrating no significant difference between the two intervention groups (MD −1.17; 95% CI −2.98 to 0.63; I ² = 97%; p = 0.20; Figure S1), despite considerable heterogeneity. Regarding efficacy outcomes, both interventions demonstrated comparable performance, with no significant differences in procedural success rate (RR 1.00; 95% CI 0.99 to 1.01; I ² = 0%; p = 0.77; Figure S1) or complete closure at the latest follow‐up (RR 1.00; 95% CI 0.98 to 1.01; I ² = 0%; p = 0.65; Figure S1). Similarly, the incidence of minor complications did not differ significantly between the two groups (RR 0.71; 95% CI 0.40 to 1.24; I ² = 0%; p = 0.23; Figure S1). In contrast to the overall analysis, however, the ASD subgroup analysis demonstrated no significant difference in the rate of major complications between the ICE and TEE groups (RR 0.66; 95% CI 0.16 to 2.78; I ² = 0%; p = 0.57; Figure S1).

3.6. Sensitivity Analysis

For procedural time, the direction and statistical significance of the pooled estimate remained largely consistent throughout most stages of the sensitivity analysis, suggesting overall robustness. However, exclusion of the study by Kim et al. [21] or Wang et al. gave rise in loss of statistical significance, indicating that these studies exerted a substantial influence on the overall estimate. In contrast, sensitivity analysis for major complications demonstrated a lack of robustness, as sequential exclusion of individual studies consistently rendered the pooled effect statistically non‐significant. For all remaining outcomes, the pooled estimates and corresponding p‐values remained stable across each iteration of the leave‐one‐out analysis, supporting the reliability of those findings. The forest plots for sensitivity analyses are illustrated on Figure 6.

FIGURE 6.

Sensitivity analysis for all assessed outcomes. (a) Procedural time. (b) Fluoroscopy time. (c) Success rate. (d) Complete closure at the latest follow‐up. (e) Major complications. (f) Minor complications.

3.7. Publication Bias

Publication bias assessment showed that both Egger's test and Begg's test yielded p‐values > 0.05 across all study outcomes, suggesting that no statistical evidence of significant publication bias was detected for the evaluated clinical endpoints. On the other hand, it is important to acknowledge that several outcomes were derived from a limited number of studies and were associated with substantial statistical heterogeneity. A comprehensive summary of publication bias analyses for each outcome is provided in Table 2.

4. Discussion

This meta‐analysis included over 5000 patients from 12 studies who underwent transcatheter closure for ASD and/or PFO, revealing several significant findings. ICE was associated with a numerically shorter procedural duration than TEE. However, this finding should be interpreted with caution owing to the substantial heterogeneity across the included studies. In contrast, fluoroscopy time was comparable between the two groups. Both imaging strategies achieved comparable procedural success rate and complete closure rates at follow‐up with negligible heterogeneity observed across these efficacy endpoints. Furthermore, a key finding was the significantly lower incidence of major complications associated with ICE, although the occurrence of minor complications was similar between the two modalities. Collectively, the available evidence indicates that ICE provides procedural efficacy comparable to TEE for guiding transcatheter closure of interatrial defects, while offering a trend toward greater procedural efficiency and the potential for improved patient safety.

ICE, a catheter‐based ultrasound introduced through femoral venous access, offers real‐time and high‐resolution visualization of intracardiac structures. This imaging modality provides a distinct advantage over TEE by facilitating the procedure under local anesthesia, contrasting with general anesthesia typically required for TEE guidance [25]. ICE is suggested to be a secure and viable alternative for guiding transcatheter closure of interatrial defects in both pediatric and adult populations [26, 27].

ICE, typically positioned in mid‐right atrial, offers significant advantages by providing excellent direct visualization of the intra‐atrial septum, defect rims, device positioning, and immediate assessment of residual shunting [28, 29]. ICE allows the interventional cardiologist to independently control the imaging, therefore streamlines workflow coordination compared to TEE, which requires a second operator and anesthesia support. Furthermore, professional communities acknowledge ICE as a recognized and acceptable alternative imaging modality for structural heart disease interventions, including transcatheter closure of ASD and PFO, particularly within experienced centers [30].

Despite the absence of statistical robustness, the observed reduction in procedural duration with ICE compared to TEE might theoretically be attributable to several interrelated factors that contribute to workflow efficiency and patient safety. Primarily, the avoidance of general anesthesia, a requirement for many TEE‐guided procedures, significantly shortens the process. As ICE is typically performed under local anesthesia, it eliminates the need for endotracheal intubation, anesthesiologist support, and associated preparation and recovery time. Thus, ICE removes critical delays and reduces overall procedural complexity [30]. Nevertheless, sedation protocols likely differed among participating centers, which may have contributed to the observed heterogeneity in procedural efficiency across studies. Furthermore, the capacity for the interventional cardiologist to control and interpret real‐time intracardiac imaging directly improves workflow. This autonomy eliminates inter‐specialty coordination between interventionists and echocardiographers, resulting in a more efficient approach to device positioning and deployment [26].

Beyond efficiency, the superior intracardiac visualization provided by ICE contributes to a reduced reliance on fluoroscopic guidance during procedural steps, thus decreasing cumulative radiation exposure. While the reduction in fluoroscopy time may appear modest, this exposure remains a clinically significant benefit, particularly for younger patients and in high‐volume centers [25]. This precise visualization also explains the lower incidence of major complications reported with ICE. Direct septal and device visualization facilitates accurate sizing and positioning, mitigating risks such as deceive embolization, erosion, or pericardial effusion [31]. Simultaneously, the avoidance of general anesthesia itself eliminates associated risks, including potential airway and hemodynamic complications, collectively contributing to a safer and more efficient procedural outcome [32].

The superior spatial resolution and real‐time intracardiac visualization afforded by ICE enable precise guidance throughout each procedural step, including wire manipulation, device positioning, deployment, and immediate evaluation of residual shunting, while substantially reducing dependence on fluoroscopic imaging. As operator expertise has advanced alongside continuous improvements in echocardiographic imaging technology, transcatheter closure of interatrial communications has progressively evolved toward minimal‐ and even zero‐fluoroscopy approaches. A recent systematic review and meta‐analysis demonstrated that zero‐fluoroscopy ASD closure achieved procedural success, complete defect closure, and safety outcomes comparable to those of conventional fluoroscopy‐guided procedures, while completely eliminating exposure to ionizing radiation [33]. These findings have important clinical implications, as cumulative radiation exposure remains a significant concern for both patients—particularly pediatric and young adult populations with a longer lifetime risk of radiation‐induced complications—and healthcare professionals, including interventional cardiologists and catheterization laboratory staff who are subjected to chronic occupational radiation exposure [34, 35]. Collectively, these data suggest that ICE should be regarded not merely as an alternative imaging modality to TEE, but as a pivotal technology facilitating the transition toward contemporary radiation‐free structural heart interventions in appropriately selected patients and experienced centers.

Despite these benefits, ICE has several limitations to consider. Firstly, ICE catheters are typically single‐use and thus represent a greater cost burden compared to the reusable TEE probes [31]. Furthermore, the specialized nature of ICE requires a distinct learning curve and a greater degree of operator expertise for proficient use. In addition, the deployment of ICE mandates a femoral venous puncture, which theoretically increases potential for vascular complications. While the pooled data from the meta‐analysis did not demonstrate a meaningful increase in minor complications, this procedural requirement remains an important consideration, particularly in pediatric populations.

Numerous limitations of this study deserve to be acknowledged. First, majority of the included studies were observational studies, and only one employed PSM. The principal source of bias in this analysis stems from residual confounding and selection bias, as most included studies employed observational designs with non‐random allocation between ICE and TEE. Several studies exhibited notable heterogeneity in baseline characteristics—particularly in mean age and defect size—which may have influenced treatment selection and consequently biased the pooled effect estimates. Second, two studies—Alqahtani et al. [16] and Kim et al. [21]—contributed substantially larger sample sizes compared to the others. Consequently, the overall pooled estimates may be disproportionately influenced by these high‐weight studies, potentially limiting the robustness of the findings. Third, significant heterogeneity and evidence of publication bias were observed for certain procedural outcomes, particularly procedural time and fluoroscopy time. The observed heterogeneity may be attributable to variability in baseline characteristics across studies, including the broad age range of participants and the inclusion of both ASD and PFO populations, which may differ in anatomical complexity and procedural requirements. Finally, the follow‐up duration varied considerably among the included studies. As a result, the reported rates of complete closure at the most recent follow‐up may not be directly comparable, given that longer follow‐up periods may allow for further device endothelialization and defect sealing. Accordingly, these findings should be interpreted cautiously, particularly when extrapolating the results to broader clinical practice.

5. Conclusion

In conclusion, overall evidence suggests that ICE serves as a safe and effective alternative imaging modality for guiding transcatheter closure of interatrial communications. Compared with TEE, ICE may offer modest procedural efficiency, as reflected by a shorter procedural duration. However, this finding should be interpreted with caution given the lack of statistical robustness. Furthermore, its use appears to correlate with a lower incidence of major complications. Both imaging modalities demonstrate comparable procedural success and complete closure rates. Overall, our findings support the feasibility and clinical utility of ICE as an alternative modality rather than establishing superiority. Nevertheless, adequately powered RCTs with more homogeneous patient populations are warranted to further clarify the comparative performance and clinical advantages of ICE.

Funding

The authors did not receive any financial support from any organization for the submitted work.

Ethics Approval

Individual clinical trials included in this meta‐analysis study were approved by the health authorities according to the local regulations and local independent ethics committees.

Consent to Participate

All participants were provided the informed consent prior to the enrolment in this study.

Consent to Publish

All of the authors have reviewed the final version of the manuscript and agreed to publish this manuscript.

Conflicts of Interest

All authors declare no competing interests.

AI Disclosure Statement

The authors used QuillBot for grammatical check and language editing only. No AI tools were used for data analysis, interpretation, or generation of scientific content.

Supporting information

Supporting information: Supplementary Materials.docx

Data Availability Statement

The authors confirm that the data supporting the findings are available within the article.