The initial treatment of any patient suspected of having a possible liver abscess is initiation of broad spectrum antibiotics. The development of broad spectrum single agents (imipenim, piperacillen/tazobactam) has replaced the traditional treatment of the combination of ampicillin, aminoglycoside, and an anaerobic drug such as metronidazole [7]. The duration of antibiotic use remains debatable, and is usually based on treatment response and the abscess characteristics.

Percutaneous drainage was first reported in 1953 but did not become accepted as standard therapy until the 1980s. It has now become the treatment of choice for pyogenic hepatic abscesses. It is usually performed either with ultrasound or CT guidance, and success rates range from approximately 60–90%. There is still some debate, however, as to percutaneous aspiration alone versus catheter drainage. Several studies have demonstrated the efficacy of percutaneous aspiration alone. Giorgio et al. reviewed 39 patients with hepatic abscesses who were treated with aspiration alone; 36 of the 39 (92.3%) were successfully treated with a single aspiration, and the other three patients only required one more aspiration. There were no deaths or complications in his study [8]. Yu et al. demonstrated a 96.8% success rate in 64 patients with aspiration alone; approximately half (49.5%) required a single aspiration and the rest of the patients required multiple aspirations. In his study, two patients died of overwhelming sepsis and another required surgical intervention for a liver laceration. However, other studies have demonstrated superiority of catheter drainage [9]. Rajak et al. randomly assigned 50 patients to aspiration or catheter drainage. Residual abscess after two aspirations was considered failure in the aspiration group, and residual abscess after catheter drainage was considered failure in the catheter group. Only 60% responded to the needle aspiration, whereas 100% responded in the catheter drainage group [10]. Zerem et al. prospectively randomized patients to percutaneous aspiration versus catheter drainage. Similar to the last study, percutaneous aspiration was successful in 67% of patients, whereas catheter drainage was successful 100% of the time [11].

Catheter drainage appears to also be successful in patients with multiloculated or multiple abscesses. A series by Liu et al. found no difference between single and multiple abscesses and had very high clinical success rates of treatment of 87% for a single abscess and 92% for multiple abscesses with catheter drainage. That study also found an 88% success rate for treatment of a single multiloculated abscess as well as a 90% success rate for multiple multiloculated abscesses [12]. Failure of catheter drainage appears to be decreasing but still exists in approximately 10% of patients. A recent case series by Mezhir et al. demonstrated only a 66% success with catheter drainage; however, in this study, 88% of patients had a history of gastrointestinal ­malignancy. Nine percent of these patients required surgical intervention, whereas the rest of the patients who failed ­percutaneous drainage died with indwelling catheters. Independent predictors of failure of catheter drainage included positive yeast cultures and communication with the biliary tree [7].

Surgical therapy is rarely necessary as the first line of intervention. If necessary, it is usually in patients with an obstructed biliary system than is not amenable to nonsurgical decompression or a ruptured abscess with sepsis. More commonly, surgical intervention is now reserved only when percutaneous drainage has failed, the abscess is not amenable to percutaneous drainage (multiloculated or large), or when there is a complication from percutaneous drainage [6].

If the cause of the hepatic abscess is unknown, a careful exploration of the abdomen should be performed to rule out any other abdominal pathology. Surgical drainage of the abscess is then performed by localization of the abscess via ultrasound or needle localization with ultrasound guidance. The abscess is then bluntly opened and the pus evacuated. Blunt finger manipulation can be used to break up loculations and adhesions. Careful hemostasis should be obtained to prevent residual fluid collections or recurrent abscess. Large bore drains are then left in place for irrigation and suction of the abscess cavity. Tan et al. retrospectively reviewed 80 patients with pyogenic abscesses >5 cm who were treated either with surgical drainage (44 patients) or percutaneous drainage (36 patients). Eighty percent of these patients had multiloculated abscesses. In this study, the surgical drainage group had less treatment failure, less secondary procedures, and a shorter length of stay. The mortality for the surgical drainage group was 4.5% and 2.8% for the percutaneous group, which was not statistically significant [13].

Some case reports have advocated primary liver resection for hepatic abscess. Hope et al. retrospectively reviewed patients with >3 cm multiloculated pyogenic abscess who were treated with percutaneous drainage along with antibiotics versus treatment with partial liver resections. The resection group had a 100% success rate of treatment and 7.4% mortality in this group, whereas the drainage group only had a 33% success rate for treatment and 4.7% mortality. Eight patients in the latter group required repeat drainage and five required surgical resection. The mortality rates between the two groups also did not reach statistical significance. The authors concluded that for large multiloculated abscesses, surgical treatment may be the primary mode of treatment of the disease [14]. Strong et al. reviewed 49 patients who underwent resection for hepatic abscesses after either failed conservative treatment or underlying hepatobiliary pathology. All of the patients had resolution of their abscesses and no patients required reoperation. The authors did report 4% mortality in their group after abscess rupture in two patients [15].