What's new?

Patent foramen ovale (PFO), present in approximately 20% to 30% of the general population, is a remnant of fetal circulation that can connect the right and left atria, potentially leading to paradoxical embolism and stroke. The unique aspects of this study include the study population selection and the analysis of demographic and clinical characteristics as well as anatomical features associated with PFO in relation to patient age. The study population included patients with a history of neurological episodes who were prescreened for PFO using transcranial Doppler (TCD) ultrasound. The criteria for this selection were: 1) PFO diagnostic workup is not performed in asymptomatic patients and is limited to those with a history of neurological events; and 2) cardiac assessment with transesophageal echocardiography is usually performed in patients with a positive TCD result. Findings of this study could offer valuable insights for establishing indications for interventional closure of PFO in different age groups.

Introduction

Patent foramen ovale (PFO) is a remnant of fetal circulation found in about 25% of the general population.^1,2 Under increased abdominal pressure, PFO may allow communication between the right and left heart chambers, potentially causing paradoxical embolization and ischemic events.^3,4 Numerous studies have linked PFO to neurological events, especially cryptogenic stroke and transient ischemic attack (TIA).^5-7 Additionally, recent discussions have highlighted its potential association with migraine.^8,9

However, given the high prevalence of PFO and the considerable heterogeneity among affected individuals, the risk of ischemic neurological events may vary depending on anatomical features, demographic factors, and clinical characteristics. This variability has led to questions about the criteria for interventional closure of PFO.^10,11

This study aims to determine patient clinical characteristics and atrial septum anatomy in relation to age and the presence of PFO, as well as to assess the factors associated with a history of stroke in patients evaluated for cardiac causes of neurological events.

Patients and methods

Results

In the study population, the distribution of neurological events was as follows: history of ischemic stroke was reported in 228 (28%), TIA in 253 (31%), and migraines in 402 (49%) patients. PFO was confirmed in 644 patients (79%) and 173 patients (21%) did not have PFO.

The patients with PFO demonstrated higher rates of TIA, syncope, and migraine history, as compared with those without PFO (Table 1). With regard to the atrial septum anatomy, those with PFO more frequently had Chiari network and ASA (Table 1).

A comparison of patients based on age is presented in Table 2. There were fewer women in the group of patients older than 45 years. Additionally, this group showed higher rates of stroke and typical cardiovascular risk factors, including arterial hypertension, diabetes mellitus, hypercholesterolemia, and varicose veins of the lower limbs. Analysis of patients based on their history of stroke is presented in Table 3. The group of patients with a history of stroke had a lower proportion of women. Furthermore, the patients in this group were older and had a higher prevalence of ASA and higher rates of arterial hypertension, hypercholesterolemia, and smoking. In Table 4, we present patient characteristics in relation to the presence of migraines. Those with a history of migraines were younger, more commonly female, and had higher rates of PFO and Chiari network.

In Supplementary material, we present analysis focused on patients 45 years old or younger in relation to the presence of PFO. Generally, the results were similar to the previous calculations presented in Table 1, which included all age groups. We reported a lower proportion of women in the PFO population, but only in patients 45 years old or younger (Supplementary material, Table S1).

Logistic regression analysis was conducted to identify predictors of past stroke episodes (Supplementary material, Table S2). Factors associated with a higher prevalence of stroke corresponded with findings presented in Table 3, and included age, male sex, arterial hypertension, hypercholesterolemia, smoking, and ASA.

Discussion

The key findings from the present study, conducted in a cohort of patients with a history of neurological ischemic events who were initially screened for PFO, are as follows: 1) individuals with confirmed PFO, regardless of age group, more frequently had a history of TIA, migraines, and syncope as well as higher rates of Chiari network and ASA; 2) there were fewer women among the patients with confirmed PFO, but only in the population of patients 45 years old or younger; 3) advanced age, male sex, typical cardiovascular risk factors, and ASA were associated with more common history of stroke.

A substantial body of evidence from numerous trials has demonstrated a significant association between PFO and ischemic neurological events.^19-24 A review of retrospective studies identified key contributors to an elevated risk of stroke in patients with PFO, including a large foramen ovale, an extensive right‑to‑left shunt, shunt at rest, increased mobility of the PFO valve, and coexisting ASA.^25,26 Furthermore, in line with findings from our study and previous research, demographic characteristics, such as older age, male sex, and typical cardiovascular risk factors also contributed to an increased risk of stroke in that patient population.^25,26

Despite the high prevalence of PFO, its exact role in the pathogenesis of stroke remains unclear. It may involve paradoxical embolization or thrombus formation within the PFO itself, which adds complexity to the understanding of its implications in stroke pathology.²⁴

In a clinical setting, identifying the source of embolic material in patients with PFO and ASA is often challenging. Despite thorough examination of the deep veins of the lower limbs and pelvis, the precise source might remain unknown.²⁷ In our study, there was no difference in the number of subjects with varicose veins between the patients with and without PFO (we only noted a trend toward a higher occurrence of this factor in the PFO cohort). Notably, varicose veins were more prevalent in the patients over 45 years of age. This suggests that in the absence of deep vein thrombosis, emboli might originate from thrombi formed at the PFO edges or on the ASA surface.²⁸ Another hypothesis is that small fibrin‑platelet fragments in the venous system may represent a potential source of paradoxical embolism, as they are too small to be detected by current diagnostic techniques. A key factor supporting this mechanism is the observed association between cerebral strokes and activities involving the Valsalva maneuver.²⁹

Previous research has indicated an association between the risk of ischemic stroke and both the size of the foramen ovale canal and the magnitude of the right‑to‑left shunt. However, standardized guidelines for quantitatively assessing contrast leakage through the PFO canal are currently lacking.³⁰ Some researchers recommend counting the number of bubbles passing through the septum as an indicator, but this modality has its limitations, including inconsistent leakage relative to PFO size and subjective bubble counting.³¹

In our study, we encountered challenges in definitively correlating PFO size with a history of ischemic stroke, primarily due to the inherent limitations in measuring PFO size during TEE. A key challenge lies in synchronizing the Valsalva maneuver with contrast administration, a procedure that rarely yields comparable images for accurate determination of PFO size between patients. Consequently, our study did not prioritize PFO dimensions. More invasive measurement techniques (eg, using a low‑pressure balloon) exist but are not routinely employed due to their complexity and extended duration of the procedure.³² Instead, we focused on evaluating shunt intensity during TEE with the Valsalva maneuver, finding that nearly half of our patients exhibited a shunt intensity corresponding to the highest degree on our semiquantitative scale.

In our study, the group with PFO exhibited a higher prevalence of ASA and other anomalies, such as the Chiari network, both of which are recognized contributors to paradoxical embolism. ASA can affect the opening and size of the PFO, thereby increasing the risk of embolism.³³ Another potential source of embolism may be fibrin‑platelet thrombi that are formed on the left atrial surface of the ASA.²⁸ Additionally, ASA is associated with paroxysmal supraventricular arrhythmias, including atrial fibrillation, which further elevates the risk of cerebral embolism.³⁴ The Chiari network, a remnant of the sinus venosus valve, also presents an increased risk of paradoxical embolism and was found to be more common in the patients with PFO in our study.^35,36

Our findings demonstrated that patients older than 45 years had a more frequent history of stroke, as compared with their younger counterparts. This contrasts with previous studies, which typically report a higher prevalence of PFO‑associated cryptogenic stroke in younger individuals.²⁴ It remains unclear whether the increased stroke incidence in the group of older patients with PFO is attributable to more frequent cardiovascular risk factors or the prolonged presence of PFO. Notably, this population also exhibited a higher prevalence of ASA, an additional recognized risk factor for stroke.

Our study observed higher rates of migraine, particularly migraine with visual aura, in the patients with PFO, consistent with previous findings.³⁷ The precise mechanism linking migraines and PFO remains unclear. One possible explanation is paradoxical microembolization, a phenomenon similar to that proposed in the pathogenesis of ischemic stroke.³⁷