Patent Foramen Ovale with Persistent Hypoxaemia and Right-to-left Shunt

Patent foramen ovale (PFO) is an important causative finding in young-onset cerebral infarction. It is also recognised in older patients with right-to-left shunt manifestations and hypoxaemia. Typically, patients with hypoxaemia due to PFO present with a history of intermittent platypnoea-orthodeoxia (POS). These findings have been reported to worsen with age due to changes in PFO morphology caused by aortic elongation, cardiac diastolic dysfunction and shape changes. Here, we report a case of successful transcatheter PFO closure with persistent right-to-left shunt hypoxaemia and multiple emboli.

Case Report

The patient was a 68-year-old man with a history of diabetes, dyslipidaemia and cerebral infarction, and he was an outpatient at a general hospital. His medications included semaglutide, dapagliflozin and cilostazol. He experienced cerebral infarction 20 years earlier and, although his detailed medical history was unknown, no apparent weakness was observed. He was admitted to the hospital as an emergency patient due to left upper abdominal pain.

Contrast-enhanced CT revealed multiple renal and splenic infarctions (Figure 1). There were no apparent pulmonary embolism and pulmonary abnormalities, including arteriovenous malformation. Initial laboratory tests upon admission revealed hepatic and renal dysfunction, as well as increasing D-dimer, and glycaemic control was poor (Table 1). No lower-limb venous thrombosis was detected. Although we started oxygen administration, the patient’s oxygen saturation remained low at approximately 80%, regardless of his body position. Cardiac function was preserved, with no atrial enlargement and no pulmonary hypertension on echocardiogram. Based on the aforementioned history, right-to-left shunt was suspected. Although the bubble test was not performed owing to poor visualisation of the area around the atrial septum caused by aortic elongation and cardiac rotation on transthoracic echocardiogram, transoesophageal echocardiography revealed PFO with persistent right-to-left shunt (Figure 2).

Aortic elongation and cardiac rotation made it difficult to evaluate the anatomy of the PFO and estimate shunt volume by echocardiography. However, persistent right-to-left shunt and a low angle between the inferior vena cava and PFO were suspected as high-risk PFO features. Furthermore, hypoxaemia was suspected to be caused by PFO right-to-left shunt. Therefore, we planned to perform right heart catheterisation and an oximetry run during PFO balloon occlusion.

We punctured the right femoral vein for intracardiac echocardiography (ACUSON AcuNav; Biosense Webster) and right heart catheterisation, and we inserted an arterial line from the right radial artery for blood gas sampling. Right heart catheterisation revealed no pulmonary hypertension and normal right atrial pressure (Table 2). We used intracardiac echocardiography to visualise the PFO and confirm the wire crossing via the PFO (Figures 3A and 3B). Transoesophageal echocardiography was not used during the procedure. The patient already showed persistent right-to-left shunt, so the bubble test was not performed. Blood oxygen saturation was increased after PFO balloon occlusion (sizing balloon II 18 mm; Abbott Cardiovascular; Table 2). The PFO size was 12 mm according to the sizing balloon measurement. Subsequently, intracardiac echocardiography-guided PFO closure was performed using a 30mm occluder device (Amplatzer Talisman, 30 mm; Abbott Cardiovascular; Figures 3C and 3D) according to the balloon sizing. Oxygen saturation normalised, and there were no other apparent complications, except for an episode of paroxysmal supraventricular tachycardia.

Following rehabilitation, the patient was discharged home and transitioned to outpatient follow-up. He was followed up at Tottori University Hospital, Tottori, Japan, for 7 months without any complaints or bleeding events, after which he was returned to his primary care physician. Antithrombotic therapy was switched from cilostazol to edoxaban after the intervention.

Discussion

We described a case of successful transcatheter closure of PFO with persistent hypoxaemia and multiple emboli. Although evidence for PFO closure in older patients has not been established and it is not generally recommended for patients with stroke, it remains a treatment option for hypoxaemia and recurrent emboli caused by right-to-left shunt.

Incidental PFO is found on <25% of autopsies.¹ As the prevalence of cryptogenic stroke due to PFO is thought to be lower in older patients, the possibility of cryptogenic stroke due to PFO has traditionally been considered in younger patients based on the Risk of Paradoxical Embolism score, which quantifies the contribution of PFO to stroke occurrence.² Current guidelines consider percutaneous PFO closure reasonable for patients who meet each of the following criteria: age 18–60 years, non-lacunar stroke, no other identified cause and high-risk PFO features.³ However, it has been reported that high-risk PFO features, such as malalignment and atrial septal aneurysm, are observed more frequently in older patients.⁴ Crescent-shaped atrial septal defects, traditionally thought of as stretched PFO, have also been reported to increase in frequency with advancing age.⁵ It has been suggested that aortic root enlargement shortens the atrial septum and increases its amplitude.⁶ These findings may result from a ‘spinnaker effect’ with inferior vena caval flow, even without an interatrial pressure gradient, opening the foramen ovale and leading to sustained right-to-left shunt.⁷ Although the patient in the present case had no thoracic spinal kyphosis, the elongated aorta compressed the atrial septum and the angle between the inferior vena cava, and the atrial septum became shallower (Figure 4). These changes may facilitate emboli from the inferior vena cava to the arterial system, regardless of the pressure gradient between the bilateral atria.

PFO-mediated hypoxaemia in older individuals is often intermittent hypoxia typified by POS. In patients with PFO-related POS and without severe pulmonary hypertension, percutaneous closure of the PFO has been considered a first-line treatment, so this approach was followed in the present case.⁸ The patient presented with position-independent hypoxaemia and had persistent right-to-left shunt in the supine position. Other cases of PFO with hypoxaemia on exertion have also been reported.⁹ It is important to consider the possibility of right-to-left shunt in patients with hypoxia that remains unexplained by other cardiopulmonary diseases.

PFO closure for the secondary prevention of cerebral infarction has demonstrated improved long-term outcomes in several randomised controlled trials, whereas the long-term efficacy of closure for hypoxaemia remains insufficiently established.^10,11 Nevertheless, case series and registry studies have demonstrated sustained symptom relief for up to 5 years, along with improved arterial oxygen saturation during upright positioning, in patients without other causes of hypoxaemia.⁸ Furthermore, transcatheter PFO closure is a safe and effective technique that immediately relieves orthodeoxia and patient symptoms.¹²

Differentiating between stretched PFO with persistent shunt and atrial septal defect can be challenging. In the present case, the absence of right ventricular enlargement, the defect location and the presence of atrial septal aneurysm led us to conclude that the patient had PFO. However, it should be noted that complex morphologies, such as atrial septal malalignment or crescent-shaped atrial septal defects, may require an atrial septal defect occluder, even in PFO.¹³

Conclusion

Right-to-left shunt due to PFO can be a differential diagnosis for persistent hypoxaemia and POS in older patients. In this case, transcatheter PFO closure was useful for preventing embolism and improving hypoxaemia.