Search

The purpose of this study was to evaluate dynamic CT findings of hepatic abscesses, especially segmental hepatic enhancement, and to clarify the cause.Twenty-four abscesses in eight patients were examined by early (30 sec) and late phase (90 sec) dynamic CT. Patients underwent abscess drainage (n = 1), hepatic resection (n = 2), or antibiotic therapy (n = 5). CT during arterial portography and CT during hepatic arteriography were performed in one patient. We retrospectively observed the frequency and changes of segmental hepatic enhancement on dynamic CT and determined its cause using radiologic and pathologic correlation.Sixteen abscesses (67%) showed transient segmental hepatic enhancement and three abscesses showed only segmental hepatic enhancement in the early phase. Four abscesses in one patient who underwent CT during arterial portography and CT during hepatic arteriography showed a segmental perfusion defect on CT during arterial portography and segmental enhancement on CT during hepatic arteriography. On follow-up dynamic CT performed 10-17 days after the initial CT, segmental hepatic enhancement surrounding hepatic abscesses decreased or disappeared in all abscesses. Pathologic examination of two patients showed marked inflammatory cell infiltration with stenosis of portal venules within the portal tracts surrounding hepatic abscesses without definite inflammation in the liver parenchyma.Segmental hepatic enhancement on dynamic CT is frequently associated with hepatic abscesses and may be caused by decreased portal flow resulting from inflammation of the portal tracts.

Results of percutaneous drainage performed in eight patients with eight liver abscesses with intrahepatic biliary communication and 22 patients with 26 liver abscesses without biliary communication were analyzed to determine whether the presence of an intrahepatic biliary communication affected the outcome of treatment. The clinical features and response to treatment of both groups were compared. The presence or absence of biliary communication was determined by injection of contrast material into the abscess under fluoroscopic guidance either during or several days after initial drainage. Duration of drainage was longer (p less than .05) in patients with communication (range, 7-44 days; mean, 22 days) than in patients without communication (range, 1-33 days; mean, 13 days). Percutaneous drainage was curative in five (63%) and palliative or temporizing in one (13%) of eight patients with communication. It was curative in 15 (68%) and palliative or temporizing in five (23%) of 22 patients without communication (p = .317). Liver abscesses with intrahepatic biliary communication did not require percutaneous transhepatic biliary diversion for cure. Despite longer duration of drainage for abscesses with intrahepatic biliary communication, the cure rates of percutaneous drainage for both groups were similar. Patients in whom an intrahepatic biliary communication was shown did not require alternative interventional or surgical measures for cure.

Percutaneous catheter drainage is now an accepted technique in the management of hepatic and other intraabdominal abscesses. Most abscesses accepted by the radiologist for drainage are unilocular with a safe percutaneous approach. This report demonstrates that even multiple hepatic abscesses may be successfully managed by percutaneous drainage. The use of computed tomography (CT) for diagnosis and sonography for safe puncture guidance is discussed.

Of 36 consecutive patients with pyogenic liver abscesses evaluated by CT, five (14%) had multiple small abscesses less than 2 cm in size. The CT appearance of the small pyogenic abscesses was compared with that of 10 patients who had either fungal or mycobacterial abscesses and with that of 50 patients who had hepatic metastases. In all five patients who had small pyogenic abscesses, the abscesses appeared to cluster, or aggregate, in a pattern that suggested the beginning of coalescence into a single, larger abscess cavity (cluster sign). This cluster appearance was not seen in in any of the patients who had fungal or mycobacterial microabscesses. It was present in only one of the patients who had confirmed hepatic metastasis. Despite the small size of the abscesses, guided needle aspiration was successful in recovering pyogenic organisms in four of the five patients. In our experience, the presence of the cluster sign suggests that the lesions are pyogenic abscesses.

Fifty computed tomographic (CT) scans in 33 patients with 37 separate episodes of hepatic abscess were reviewed retrospectively. Abnormalities were detected in all but one case (97% sensitivity). However, a characteristic CT appearance of hepatic abscess was not evident. Most intrahepatic foci were solitary, but multiple abscesses were present in 34% of cases. The CT appearance of the lesions varied from well defined, rounded cavities with contents near water density, resembling poorly defined hepatic cysts, to higher-density foci indistinguishable from hepatic neoplasms. In only 19% was gas present within the abscess. Rim enhancement was seen in only 6%. This study suggests that CT is a sensitive test for detecting hepatic abscess but is often nonspecific.

Sonography has proven useful in the detection of intmaabdominal abscesses [1 ]. Although used for this purpose for some time, it has only recently been appreciated that these lesions may have varied and perhaps confusing sonogmaphic appearances. While the typical abscess is usually anechoic [2], theme are reports of increased echogenicity in some lesions [3-5]. Kmessel and Filly [4] reported gas-containing abdominal abscesses appearing as densely echogenic masses and postulate a microbubble effect for the marked echogenicity. We describe a case of multiple hepatic abscesses with marked echogenicity that did not contain gross gas.