Arterioportal fistulae are usually silent clinically until complications of portal hypertension occur. \Ve present the case of a 60-year-old man with an abdominal bruit appearing 4V2 years after a splenectomy and aortic aneurysm repair and suspected to be caused by an aorto-caval 6stula. Pulsed Doppler ultrasonography detected a large hepatic arteriovenous 6stulu, confirmed hv dvnumic computed tomography and angiography. •
REPORT OF A CASEA 60-year-old man underwent aorto-hifcmoral bypass for a saccular aneurvsm of the abdominal aorta. At the time of surgery, splcncctomy was performed to correct hypersplcnism and two needle biopsies were ohtaim:!d from the lel'i: lobe and a wedge biopsy from the right Johe of the liver to rule out a clinically suspected lymphoma. The inforior vena cava was laccmtecl and repaired. There were no immediate postoperative compli• cations . Four and a half years later, he developed c; c -ardiac failure. A right upper quadrant abdominal bruit was heard. Heal -time ultrasound examination disclosed a normal inferior vena cava and un aneurvsm of the abdominal aorta. The left branch of the 1;ortul vein was grossly dilated as was what was presumed to he and, at Doppler investigation, proven to be the left hepatic artery! ( fig. 1).A fistula between the left hepntic nnd the left portal vein was suspected. Pulsed Doppler examination showed greatly turbulent high velocity, bidirectional, systolodiastolic flow at the site of the fistula {fig. lB) and a normal direction of the hlood flow in the remainder of the portal system .IiThere was no ascitcs. No other signs of portal hypertension were seen .Dynamic computed tomography showed simultaneous opacification of the left hepatic arter)' and intrahcpatic portal vein ( fig. 2). There was an increased "'onccn trntion of' contrast in the left hepatic Johe . e c onda ry to prcfcrentiul left hlood supply {fig. 28).Abdominal aortography confirmed the intrahepatic ar· tcriovenous fistuln.
DISCUSSIONHepatoportal 6stulae are usually the result of rupture of a pre-existing hepatic artery aneurysm, 3A trauma;' or previous biopsy.s In our pa- tient the left hepatic needle biopsies performed 4V~ years earlier probably caused the fistula. Other causes include tumor invasion of the two vessels, cirrhosis. congenital arteriovenous malformations, 3 and various transhepatic manipulations procedures. n.fl These do not apply in this case. An aortognun done prior to the aneurysm repair lmd shown normal hepatic arteries.Our patient had not yet developed the most common presenting clinical foatures of in trahepatic arterio-portal fistula, those of portal hypertension. 3 ·fl He presented with high output cardiac failure, a relatively rare complication of this type of fistuln.; At the time of discovcrv of his abdominal bruit, the latter was assumed to ~l ri se from an iatrogenic :wrtocuval fistula . The ultrasonographic find ings were thus a surprise. The Doppler results are classic for an artcriovenous fistula: bidirectional systolodiastolic flow at the site of the fistu...