A 67-year-old woman previously diagnosed with hypothyroidism with no history of liver disease or trauma presented with memory impairment of approximately one-month duration and behavioral changes that had begun three days previously. The patient presented with mild confusion on examination. Laboratory tests revealed an abnormally high ammonia level (149 mmol/L). After treatment with a Duphalac enema, her ammonia level dropped to 34 mmol/L.
Brain magnetic resonance imaging (MRI) showed high signal intensity in the globus pallidus of the basal ganglia and the cerebral peduncle of the midbrain on T1-weighted imaging, suggestive of hepatic encephalopathy (). Computed tomography (CT) scanning of the abdomen revealed two portovenous shunts involving an aneurysmal sac between the right portal and right hepatic veins with details as follows: a large shunt tract (8-9 mm in diameter, 5 cm in length) communicating from the central portion of the right posterior portal vein to the aneurysmal sac (3.6 × 2.5 cm) and a small shunt tract (4 mm in diameter, 3 cm in length) communicating directly from the right anterior inferior branch to the aneurysmal sac ().
67-year-old woman with intrahepatic portosystemic venous shunt (IPSVS).
Based on these findings, the cause of the patient's encephalopathy was diagnosed as IPSVS, and shunt embolization was determined to be the appropriate treatment. An ultrasound-guided puncture of the right internal jugular vein was initially performed in preparation for a transjugular transvenous approach to embolization. However, the 5-Fr catheter (Cobra®; Cook, Bloomington, IN, USA) could not be safely advanced to the main portal vein due to kinking and distortion of the catheter in the aneurysmal sac. We were concerned about laceration of the aneurysmal sac due to increasing pressure on the sac wall; therefore, we changed to the transhepatic approach.
To access the portal system, ultrasound-guided transhepatic direct puncture of the right posterior branch of the portal vein was performed using a 21-gauge needle through the right intercostal space, while a 0.018-inch wire was inserted into the portal vein trunk. A pair of coaxially mounted catheters was advanced over the 0.018-inch wire; after exchange to a 0.035-inch guide wire, a 6-Fr sheath (Flexor Check-Flo Introducer®; Cook, Bloomington, IN, USA) was finally inserted into the portal vein along the guide wire. A 5-Fr catheter (Cobra®; Cook, Bloomington, IN, USA) was inserted into the main portal vein for performing direct portography and measuring portal vein pressure. The mean portal venous pressure was 13 mm Hg. A direct portogram confirmed the presence of IPSVS, revealing two shunt tracts from the portal vein (). We measured the size and length of the two shunt tracts from a right oblique view using the measuring program of the angiographic equipment. Based on the CT and angiographic findings, we decided to use coils for the embolization of the small tract and AVP II for the embolization of the large tract.
We performed embolization of the small shunt tract by inserting a 5-Fr catheter into it through the anterior portal vein via the transhepatic route; embolization was then performed using three coils (Nester® embolization coil; Cook, Bloomington, IN, USA) (). We performed embolization of the large shunt tract was performed by carefully advancing a 6-Fr long sheath into this shunt tract along a guide wire and then deploying the AVP II (12 mm in diameter) from the aneurysmal sac side of the tract. We first placed three consecutive AVP II devices to reduce procedure time and increase placement precision and then waited for 12 minutes. However, portography showed persistent flow into the aneurysmal sac, and additional AVP II devices were placed into the remnant tract. Complete occlusion of the large shunt tract was ultimately achieved with the use of five AVP-II devices.
Post-embolization portography via a 5-Fr catheter revealed the absence of blood flow through the shunt, confirming complete occlusion. Measurement of portal venous pressure after embolization revealed markedly increased pressure in the main portal vein from 13 mm Hg to 27 mm Hg. After occlusion was confirmed, eight coils were carefully inserted into the sheath's parenchymal tract to prevent massive bleeding ().
Three days after the procedure, laboratory tests indicated that the patient's serum ammonia level had decreased to within the normal range (), while physical examination revealed complete remission of all symptoms related to hepatic encephalopathy. The patient was discharged four days after the procedure with no complications. Follow-up CT scanning three months after the procedure indicated occlusion of the IPSVS and collapse of the aneurysmal sac (). Upon patient examination and consultation, we found no symptoms of hepatic encephalopathy or portal hypertension.