The foramen ovale is a vital fetal structure that allows oxygenated blood to bypass the fetal lungs and flow directly from the right to the left atrium. After birth, as the lungs start functioning, increased left atrial pressure usually causes the foramen ovale to close. However, in 10–25% of people, closure is incomplete, leading to a patent foramen ovale (PFO).1–3 While many individuals with PFO remain asymptomatic throughout their lives, some may experience paradoxical embolism, potentially causing ischemic events such as stroke.4–7 Cryptogenic stroke, defined as a stroke of unknown origin after extensive diagnostic evaluation, accounts for a significant portion of ischemic strokes, particularly in younger patients.8–10 Despite advancements in stroke prevention and management, cryptogenic stroke continues to present a diagnostic and therapeutic challenge.11–15

The potential link between PFO and cryptogenic stroke has been the subject of considerable debate. Some studies suggest a strong correlation, particularly in younger individuals without traditional risk factors, while others argue that PFO may often be an incidental finding.16-21 In older patients with common stroke risk factors like hypertension or diabetes, the role of PFO is even less clear.19

The possibility of paradoxical embolism has led to the consideration of PFO closure as a treatment option for preventing recurrent strokes. Percutaneous closure, a minimally invasive procedure, has emerged as a preferred alternative to medical therapy, particularly in patients at a high risk of recurrent stroke.

Several randomized controlled trials (RCT) and observational studies have evaluated the efficacy and safety of percutaneous PFO closure in patients with cryptogenic stroke. Early RCT – CLOSURE I, PC Trial and RESPECT – did not demonstrate a definitive benefit of PFO closure for secondary stroke prevention.22–24 These inconclusive results tempered enthusiasm for PFO closure as a strategy to prevent recurrent strokes. However, in March 2016, a meta-analysis of patient-level data from three trials was published, providing a more nuanced understanding of the potential benefits of PFO closure.25 This meta-analysis revealed that PFO closure was more effective than medical therapy in reducing the risk of recurrent ischemic stroke. The benefit of PFO closure was even more pronounced when the analysis focused solely on trials using the Amplatzer PFO occluder device (PC Trial and RESPECT). These findings suggest that, particularly with certain closure devices, PFO closure may offer a significant advantage in preventing recurrent strokes in patients with cryptogenic stroke.

Further evidence supporting PFO closure emerged from subsequent trials. The REDUCE trial found that PFO closure was superior to antiplatelet therapy in preventing recurrent ischemic strokes among patients with cryptogenic stroke.26 Similarly, the CLOSE trial demonstrated that PFO closure was more effective than both anticoagulation and antiplatelet therapy in reducing the risk of recurrent stroke in patients with cryptogenic stroke and high-risk PFO characteristics (atrial septal aneurysm or large right-to-left shunt).27 The long-term follow-up data from the RESPECT trial, with a median follow-up of 5.9 years, demonstrated a significant reduction in stroke recurrence in patients who underwent PFO closure, compared to those who received medical therapy (either antiplatelet or anticoagulant).28 DEFENSE-PFO trial further reinforced this, showing that in patients with high-risk anatomical PFO (atrial septal aneurysm [ASA], hypermobility, or large size), percutaneous closure significantly lowered the risk of recurrent strokes compared to medical therapy (antiplatelet therapy or anticoagulation).29 Together, these trials highlight PFO closure as a superior strategy for preventing stroke recurrence in selected patients.

Optimizing outcomes of PFO closure in cryptogenic stroke requires careful patient selection, considering both clinical and anatomical factors. Younger patients, particularly those under 60 with fewer stroke risk factors, are typically better candidates.30,31 Patients with a history of recurrent cryptogenic strokes despite medical therapy or those with venous thrombosis are also prime candidates for PFO closure.31–33 Anatomical characteristics of the PFO play a crucial role in determining the need for closure. Larger PFOs with significant right-to-left shunting increase stroke risk and may benefit more from closure.5,34–38 The presence of an ASA, often found alongside PFO, further elevates embolism risk and may warrant closure.5,34,36–38 Other structures, such as the Eustachian valve and Chiari's network, can also contribute to embolism risk or complicate catheter-based interventions, influencing the decision to close the PFO.37–39 Scoring systems and risk stratification tools aid in this selection. The Risk of Paradoxical Embolism (RoPE) score estimates the likelihood of a cryptogenic stroke being related to a PFO; higher scores indicate greater benefit from closure.31 The PFO-Associated Stroke Causal Likelihood (PASCAL) Classification further stratifies patients into “probable,” “possible,” and “unlikely” categories based on clinical, anatomical, and imaging data.40

A meta-analysis of six randomized clinical trials involving 3740 patients demonstrated that the risk reduction for recurrent stroke with PFO closure varied depending on the estimated likelihood that the stroke was causally related to the PFO.41 The relative risk reduction of recurrent ischemic stroke following PFO closure was 90% in patients in the “probable” PASCAL category, 62% in the “possible” category, and nonexistent in the “unlikely” category. Moreover, the difference in safety outcomes between the device and medical therapy groups was consistently greater in the “unlikely” group compared to the “probable” or “possible” groups. These findings underscore the importance of precise patient selection, as the benefit of PFO closure is most pronounced in those with a higher likelihood that their stroke is causally linked to the PFO.

The Society for Cardiovascular Angiography & Interventions guidelines42 now provide a structured framework based on the latest evidence. They offer a strong recommendation (moderate certainty) for percutaneous PFO closure over antiplatelet therapy in patients aged 18–60 with prior PFO-related stroke–regardless of anatomical features – with a RoPE score ≥7 identifying those likely to gain the greatest benefit. For patients >60, PFO closure is conditionally suggested (very low certainty). The guidelines also advise against routine closure in scenarios such as atrial fibrillation, transient ischemic attack (TIA) without stroke, thrombophilia in the absence of stroke, decompression illness, migraine, and systemic embolism without prior stroke (all conditional with very low certainty).

Despite the growing body of evidence supporting percutaneous PFO closure in patients with stroke, significant gaps in knowledge remain, particularly concerning older patients (>60 years), presenting with TIA or systemic embolism (without stroke) and patients with thrombophilia. Furthermore, most existing data are derived from multicenter studies or registries, which may not fully capture the nuances of individual center practices or patient populations. Thus, there is a need for additional data from single-center experiences to provide a more consistent understanding of the clinical outcomes, procedural success rates, and complication profiles associated with percutaneous PFO closure in routine practice.

Our article aims to report our center's experience with the percutaneous closure of PFO in patients presenting with cryptogenic stroke. We provide detailed analysis of our patient selection criteria, procedural techniques, and outcomes. Additionally, we explore the incidence of periprocedural complications. By sharing our experience, we aim to contribute to the growing body of evidence on percutaneous PFO closure and provide insights that may help refine patient selection and optimize procedural outcomes.

This was an observational retrospective single-center study. It included all the patients referred for a cardiology and neurology group consultation following a presumed cryptogenic stroke with a diagnosis of PFO and referred for percutaneous PFO closure between the 1 February 2021 and the 29 February 2024 at our center.

Of the 62 patients referred for a cardiology and neurology group consultation after a presumed cryptogenic stroke with a diagnosis of PFO, 32 (51.6%) were deemed to have a strong indication for percutaneous PFO closure (Figure 1). In the remaining patients, taking into account individual clinical and structural characteristics, it was considered that there was not enough evidence to support the procedure. Some of the reasons for not proceeding were low RoPE score and/or low PASCAL score, dubious clinical presentation and lack of a clear stroke image on CT and brain MRI, image suggestive of small vessel disease, among others. The demographic and clinical characteristics of our population are detailed in Table 1, which includes both the overall cohort initially proposed for evaluation as well as a comparison between patients selected for PFO closure and those who were not.

Figure 1.

Flow diagram of patient selection.

Regarding the 32 patients who underwent percutaneous PFO closure, they were predominantly male (20, 62.5%), with a mean age of 45.7±9.2 years.

The functional and anatomical characteristics of PFOs are described in Table 2. Most were considered to be high risk (27, 84.4%) and 37.5% were considered complex PFO. The mean RoPE score was 7.5±0.98. The majority of patients were classified as PASCAL “probable” (23, 71.9%) and the rest as PASCAL “possible” (9, 28.1%).

All procedures were performed under general anesthesia and guided by both TEE and fluoroscopic monitoring.

Amplatzer™ Talisman™ PFO Occluder Abbott Cardiovascular USA with a diameter of 25 mm was used in most patients (24, 75.0%), followed by Amplatzer™ Cribriform Occluder Abbott Cardiovascular USA with a diameter of 25 mm (4, 12.5%) and Amplatzer™ Talisman™ PFO Occluder Abbott Cardiovascular USA with a diameter of 30 mm (3, 9.4%). In one patient, the anatomical characteristics of the PFO – namely, a very short distance from the defect to the aorta and aortic root dilation – precluded the use of an occlusion device. As a result, percutaneous closure was performed using the Noble Stitch™. Despite this procedure, a significant residual shunt remained. However, after the first procedure, there was enough aortic edge to subsequently implant an Amplatzer™ Talisman™ PFO Occluder Abbott Cardiovascular USA device, which successfully eliminated the residual shunt.

Acute success of the percutaneous PFO closure was achieved in all 32 patients (100%), as confirmed by TEE with no evidence of residual shunt lateral to the device.

There were no immediate significant periprocedural complications, and no major complications such as death, cardiac tamponade, device embolization, or serious bleeding were observed during the surveillance period. Regarding minor complications, two patients had a hematoma on the femoral access requiring manual and mechanical compression, with no additional complications related to the venous access. No periprocedural or post-procedural arrhythmias were observed.

All patients were discharged within 24 hours following the procedure after a transthoracic echocardiogram (TTE) confirmed the absence of complications.

All patients maintained follow-up for at least 12 months, with bubble echocardiography at 6 and 12 months confirming complete PFO closure in all cases.

Our results demonstrate that percutaneous PFO closure is both highly effective and safe in a carefully selected patient population. The acute procedural success rate was 100%, with no significant periprocedural complications – such as cardiac tamponade, device embolization, or serious bleeding – observed, corroborating findings from previous studies and clinical trials.26,27

These outcomes underscore the importance of appropriate patient selection and meticulous procedural planning, including the use of advanced imaging techniques such as TEE in addition to fluoroscopy for device deployment. The absence of major complications and the low incidence of minor complications, such as femoral access site hematomas, highlight the procedural safety profile. This aligns with the safety data from other RCT and registries, reinforcing that percutaneous closure, when performed in experienced centers, is associated with a low risk of adverse events.25–27 Our study population predominantly comprised younger patients with a mean age of 45.7 years and high RoPE scores, indicating a higher likelihood of a causal relationship between PFO and cryptogenic stroke. A high RoPE score (mean 7.5) among patients who underwent PFO closure suggests that these individuals were appropriately selected based on a high pre-test probability of stroke recurrence due to paradoxical embolism. This patient profile is consistent with current guidelines that recommend PFO closure in younger patients with fewer traditional stroke risk factors and recurrent cryptogenic strokes, despite optimal medical therapy.28,29 Moreover, most of the patients had high-risk anatomical features, such as a large PFO or an ASA, which further justifies the decision for closure. These anatomical characteristics are known to increase the RoPE, making closure a more favorable option to prevent recurrent strokes.5,32–36 The favorable outcomes observed, with a 100% closure rate at six months post-procedure, further substantiate the effectiveness of this intervention.

Regarding the procedural technique, it is likely that the use of general anesthesia and intraprocedural imaging contributed to the high success rates and low complication rates observed in our study. Our results emphasize the importance of a detailed procedural approach in optimizing outcomes for patients undergoing percutaneous PFO closure. However, general anesthesia increases procedural complexity, requires dedicated Anesthesiology support, prolongs recovery time, and raises overall costs. It may also limit scheduling flexibility in centers with constrained anesthesia resources and carries additional risks, particularly in patients with comorbidities. These factors must be considered in procedural planning and patient selection in other centers.

The need for general anesthesia is due to the discomfort related to conducting the TEE-guided procedure. Fluoroscopy and nasal TEE offer promising alternatives to traditional TEE, particularly in minimizing the need for general anesthesia. Studies have demonstrated that fluoroscopy, while less detailed than TEE, provides adequate guidance for PFO closure with a low incidence of significant complications. It enables effective procedural monitoring, while reducing the risks and costs associated with general anesthesia.42,43 Nasal TEE, using a thinner probe, has been shown to be a viable alternative to conventional TEE, offering high-resolution imaging with less patient discomfort and potentially eliminating the need for general anesthesia.44,45 Both approaches have demonstrated favorable safety profiles, including lower rates of procedural complications and improved patient comfort, making them valuable options for centers seeking to optimize efficiency and minimize anesthesia-related challenges. The adoption of these techniques could expand procedural accessibility, though centers must carefully consider trade-offs, including imaging quality, operator experience, and resource availability.42,44

Additionally, intracardiac echocardiography (ICE) has emerged as an increasingly utilized imaging alternative for PFO closure, with several potential advantages.46 ICE provides high-resolution, real-time intracardiac imaging and typically enables the procedure to be performed under conscious sedation rather than general anesthesia, thereby reducing anesthetic exposure, recovery time, and logistical constraints. Studies have reported comparable procedural success and safety outcomes between ICE-guided and TEE-guided closure, with the added benefit of shorter fluoroscopy times and lower radiation exposure in some series. These advantages can translate into improved efficiency and greater accessibility, particularly in centers with limited anesthesia resources. Nevertheless, ICE also has shortcomings, including the cost of disposable catheters, the need for additional venous access, and a learning curve for image acquisition and interpretation. Furthermore, while ICE provides excellent local visualization, it does not replicate all TEE planes, and in complex anatomies, TEE may remain the preferred option. Thus, the choice of imaging modality should be individualized, balancing patient comfort, procedural complexity, institutional expertise, and resource availability.

While our findings support the use of percutaneous PFO closure in appropriately selected patients with cryptogenic stroke, several limitations must be acknowledged. First, the study's observational and retrospective design inherently limits the ability to draw definitive causal inferences. Moreover, the study was conducted at a single center, which may limit the generalizability of the results to other settings with different patient populations or varying levels of procedural expertise. Additionally, the relatively short follow-up period precludes conclusions about the long-term durability of PFO closure and its impact on term recurrent stroke risk. Future research should aim to provide more extended follow-up data to assess the longevity of the closure efficacy and the potential for late complications. Given the ongoing debate over the pathogenic role of PFO in cryptogenic stroke, randomized trials with larger sample sizes and stratified patient populations are needed to establish more definitive guidelines.

Our findings underscore the importance of a multidisciplinary approach in managing patients with cryptogenic stroke and PFO, involving neurologists, interventional cardiologists and imaging-based cardiologists. A key strength of our organizational model is the detailed discussion of each patient in a group setting. This collaborative review ensures that every case is thoroughly evaluated, allowing us to refine the selection criteria and focus on patients most likely to benefit from PFO closure. By narrowing down the selection grid, we can target the intervention more effectively, improving outcomes and minimizing unnecessary interventions. While PFO closure can be considered in other contexts beyond cryptogenic stroke, such as in patients with migraines, platypnea-orthodeoxia syndrome, TIA, or systemic embolization, the supporting evidence for these indications is limited, and recommendations for closure in these cases remain conditional.

As highlighted by a recent report, many patients undergoing PFO closure may not meet the most stringent criteria, raising concerns about overuse of the procedure in settings where a thorough pre-procedural evaluation might be lacking.47 In our model, careful patient evaluation, adherence to standardized procedural protocols, and the use of advanced imaging techniques such as TEE, in addition to fluoroscopy, contribute to high success rates and low complication incidences. The acute procedural success rate of 100%, without significant periprocedural complications, aligns with other studies and reinforces the safety of the approach when performed by experienced teams. Additionally, our patient population, consisting primarily of younger individuals with high RoPE scores and high-risk anatomical features like large PFOs or ASA, justifies PFO closure. These findings suggest that in carefully selected patients, percutaneous closure is a highly effective and safe intervention for reducing the risk of recurrent stroke. Nevertheless, further research into long-term outcomes and larger randomized trials are needed to validate these findings across broader populations.

Building on our single-center experience, future research should aim to prospectively validate our patient selection criteria and procedural protocols in larger multicenter cohorts. Longer-term follow-up studies are needed to evaluate the durability of PFO closure, late complications, and the sustained impact on recurrent stroke risk beyond one year. Additionally, exploring less invasive procedural strategies, such as the use of TTE, ICE, or nasal TEE instead of general anesthesia-guided TEE, may help optimize procedural safety, reduce costs, and expand accessibility across different clinical settings. Collecting standardized data on these approaches will provide valuable insights into procedural efficiency, patient comfort, and long-term outcomes, helping to refine best practices for PFO closure in diverse populations.

In conclusion, our study reinforces the role of percutaneous PFO closure as an effective and safe intervention in selected patients with cryptogenic stroke. The findings highlight the importance of a tailored approach that incorporates both clinical and anatomical factors to optimize patient and procedural outcomes. Continued research is warranted to further refine these criteria and explore the long-term benefits of PFO closure in diverse patient populations. Nonetheless, the retrospective single-center design, short follow-up period, and predominance of younger patients with high RoPE scores limit the generalizability and long-term applicability of our findings. Reliance on general anesthesia for TEE guidance may also reduce its relevance to centers using alternative imaging strategies. Larger prospective studies with an extended follow-up period are needed to confirm the durability of closure, better define patient subgroups most likely to benefit, and evaluate less invasive procedural approaches.

No funding sources were used in this research.

The authors have no potential conflicts of interest to disclose.