Discussion In the case of a right-to-left shunt, venous thrombi can pass from the venous to the arterial circulation. Although PFOs are common (25-30% in a healthy population), cryptogenic strokes among patients with PFO are rare.[1-3] This can be explained by the hypothesis that the PFO functions like a valve, only allowing flow when the pressure difference between the right and left atrium is positive. In this case, the venous thromboembolism was most likely caused by immobilisation due to a fracture. It developed despite the use of thrombosis prophylaxis. Although thrombosis prophylaxis does not offer full protection from venous thromboembolism, the occurrence of a thromboembolic event during prophylaxis increases the risk of recurrence. The patient being overweight increased this risk even further, thus lifelong rivaroxaban was prescribed. However, the question remains if the PFO should be closed in addition to anticoagulant therapy. A recent position paper published in 2019 summarises most of the reviews and meta analyses recently published on this topic. It states that a clinician must consider two things to determine the course of treatment: 1) the probability that the PFO has a relevant role in the observed clinical picture; 2) the likelihood that the observed clinical event will recur.[6] Given the presence of deep vein thrombosis, the young age of the patient and the observed elevated right atrial pressures with a permeable PFO, the probability that the embolism was due to the PFO is considered high. This is supported by the RoPE index of 62%. The likelihood of recurrence is more difficult to estimate. Recurrence of a paradoxical embolism is certainly possible if the PFO is not closed. However, the venous thrombosis was most likely caused by immobilisation due to a fracture, which is transient. Therefore, recurrence risk is clearly reduced once the patient is mobile again. Moreover, the chance of new venous thromboembolic events is reduced by anticoagulant therapy. In addition, elevated right atrial pressure and aneurysmatic atrial septum were no longer present at follow-up, which reduces the chance of a right-to-left shunt occurring through the PFO. In this case, the position paper does not make a clear statement about whether the PFO should be closed. It does say that medicinal therapy should be considered. Multiple studies find percutaneous PFO closure to be superior to medicinal therapy alone in preventing paradoxical embolisms and show a relative risk reduction ranging from 38.0% to 50.5%.[7-12] However, most studies compared percutaneous PFO closure to antiplatelet therapy alone instead of anticoagulants. The studies that did compare PFO closure with the use of anticoagulants did not find convincing evidence that PFO closure was beneficial.[12-14] Moreover, the risks of percutaneous PFO closure need to be taken into account when deciding on the course of treatment. Multiple studies report an increased risk of new-onset atrial fibrillation, although mostly transient, with an incidence of 4.26% after percutaneous PFO closure versus 0.67% in patients being treated with medicinal therapy alone. In addition, patients developed complications related to the implantation of the device, such as device thrombosis (1-2%) and pericardial effusion (0.5-1%).[6,12] Finally, closing a PFO requires the use of double platelet aggregation inhibitors for three to six months, which is associated with an additional risk of bleeding. Despite these risks, the all-cause mortality rates between both groups are similar (RR 0.73, p=0.42).[8,10,11,15] In this case, because rivaroxaban was prescribed, the additional risk reduction achieved by closing the PFO is limited and might not justify the risks associated with percutaneous closure. In consultation with the patient and a multidisciplinary team, it was decided not to close the PFO.

Conflict of interest

All authors declare no conflict of interest. No funding or financial support was received.