Background and purpose: The right-to-left shunt from pulmonary arteriovenous fistula (P-AVF) with Rendu-Osler-Weber (R-O-W) disease can cause paradoxical brain embolism. However, it has remains unclear whether the isolated P-AVF without Rendu-Osler-Weber (R-O-W) disease is associated with ischemic stroke, in particular, paradoxical brain embolism. Our group previously reported a case with paradoxical brain embolism associated with isolated P-AVF without R-O-W disease in 1996. Furthermore, in 1999, we reported that transcranial Doppler (TCD) with saline contrast medium was useful for identifying the presence of P-AVF as a right-to-left shunt. Therefore, when we have an embolic stroke patient with unknown source, we have performed TCD for detecting P-AVF since 1998. The aim of this study was to investigate the frequency of brain infarction associated with isolated P-AVF without R-O-W disease and to evaluate clinical characteristics including a previously reported case, and to elucidate the stroke mechanism.
Method and subjects: We studied patients with brain infarction associated with isolated P-AVF without R-O-W disease from 642 ischemic stroke patients admitted to our group within 7 days of stroke onset between August 1998 and May 2002.
Results: Four patients (0.6%) was diagnosed as having brain infarction associated with isolated P-AVF without R-O-W disease. Clinical characteristics of 5 patients were as follows; 1) all patients were middle aged women (mean, 61 year old), 2) they had no evidences of chronic hypoxia such as cyanosis, dyspnea, and erythrocytosis, 3) all strokes occurred in the morning, 4) the diagnosis of brain embolism was made by routine neuroimaging studies, but no potential cardiac nor arterial sources of emboli were detected, 5) the location of brain infarcts on MRI was more frequently in vertebrobasilar circulation (4 cases) than in carotid circulation (1 case), 6) all patients had pulmonary embolism and 4 suffered from deep venous thrombosis, 7) all patients had a history of brain infarction or transient ischemic attack (TIA) before the present attack, 8) all patients had a single P-AVF and its location was right lower lobe in 4 patients and the left lower lobe in the remaining one patient, and 9) all patients could be treated with catheter embolization of P-AVF without recurrent stroke for 37 months in average (range 2-100 months).
Conclusion: The isolated P-AVF without R-O-W disease can cause paradoxical brain embolism. Catheter embolization of P-AVF may be effective in prevention for recurrent stroke. We should not overlooked isolated P-AVF as a right-to-left shunt in embolic stroke patients with unknown etiology.