Pyogenic hepatic abscesses typically originate from enteric infectious sources (appendicitis and diverticulitis), biliary sources, or via hematogenous spread from endocarditis or poor dentition. Evaluating pyogenic liver abscess from the years 1994–2005, investigators in the United States found the incidence to be 3.6 per 100,000 admissions. Additionally, they found an in-hospital mortality of 5.6 % with mortality being associated with older age, those who did not have private health care insurance, and medical comorbidities (cirrhosis, chronic kidney disease, and cancer) [1]. A similar investigation in China, over the years 1995–2008, showed the mean age of patients was 57 years, 59 % were male, and the case fatality rate was 10 %. The most common etiology of hepatic abscess was cholecystitis and/or cholangitis (34 %) and the following factors were associated with increased mortality: Acute Physiology and Chronic Health Evaluation (APACHE II) scores ≥15, Simplified Acute Physiology Score (SAPS 2) scores ≥28, gas within the liver abscess, and infection with anaerobes [2].

Compared to pyogenic liver abscess, amoebic liver abscess tends to be less common and appears to be decreasing in incidence. Amebiasis is derived from infection with the anaerobic protozoan E. histolytica after infection of the colon and subsequent invasion of the portal venous system. Patients with amebiasis typically present with fever, malaise, and right upper quadrant pain. Utilizing the United States Nationwide Inpatient Sample from 1993 to 2007, Congly et al. showed an overall incidence of hospitalization of 1.38 per million admissions, with a decreasing incidence over time during the study period. Eighty-one percent of patients were male, approximately half were of Hispanic origin, and one-third had private health insurance. The mortality in the study was 0.8 % with predictors of mortality being age greater than 60 years, female gender, and those without private health insurance [3].

Fungal abscesses are also relatively uncommon and typically present in the immunocompromised host. Over a 20-year period at The Johns Hopkins Hospital, only 8 solitary fungal and 34 mixed fungal/bacterial abscesses were identified. Biliary tract disease was much more likely in patients with combined fungal/bacterial abscesses, with 74 % of those having prior biliary surgery or indwelling biliary stents. All of the patients in the pure fungal abscess category were immunocompromised, all having received chemotherapy. Purely fungal abscesses were smaller than the mixed type [4].

The majority of pyogenic liver abscesses is derived from enteric or biliary sources, thus the microbiology is frequently polymicrobial in nature [5]. However, given advances in antimicrobial therapy and imaging techniques, abscess formation from pylephlebitis is decreasing in incidence in the current era. In a 2010 population-based study from the United States, streptococcal species and Escherichia coli were the most common bacterial species isolated [1]. Increasingly, Klebsiella pneumonia is being a reported isolate from hepatic abscess growth, with initial reports from Taiwan [6]. A 2014 study from Texas showed Klebsiella species to be the most common isolate from a series of 49 patients with hepatic abscess. Additionally, in these Klebsiella-infected patients, they found a significant number of associated malignancies [7]. Unilocular K. pneumonia liver abscesses are at increased risk of having associated septic pulmonary emboli and other extra-pulmonary metastatic infections [8].

The vast majority of liver abscesses can be treated with antimicrobial therapy directed at the appropriate pathogen and percutaneous drainage with either ultrasound or CT guidance [9]. Initial therapy begins with broad-spectrum antibiotics directed at common pathogens. Culture data should be obtained from the abscess itself by aspiration with possible placement of a drainage catheter. Antibiotic therapy can then be tailored depending on culture data that return. Factors shown to be associated with failed medical treatment include the presence of a gas-forming abscess and shock present on admission [10].

Surgical therapy remains an important consideration in selected patients. Investigators in Rhode Island devised a treatment algorithm for hepatic abscesses <3 cm, unilocular hepatic abscesses >3 cm, and multilocular hepatic abscesses >3 cm. They were able to show 100 % treatment success with antibiotics alone for those with the small abscesses <3 cm. Similarly, they had 100 % treatment success with surgical therapy of the multilocular abscesses >3 cm. Percutaneous therapy combined with antibiotic therapy provided mixed results for patients with unilocular abscesses >3 cm and multilocular abscesses >3 cm [11].

Tan et al. examined two populations of patients with hepatic abscesses greater than 5 cm, one group treated with surgical and one with percutaneous drainage. They found that patients managed with surgical therapy had fewer treatment failures, required fewer procedures, and had shorter hospital lengths of stay. Additionally, they found no difference in morbidity or mortality between the two groups [12].

In another series, investigators in China studied 31 patients undergoing simultaneous treatment of biliary disease and hepatic abscess. Patients were managed with either laparoscopic or open surgical techniques. They found no difference in operating time, blood loss and transfusion rates, postoperative morbidity, or abscess recurrence between the two groups. Patients managed with the laparoscopic approach had a faster return of bowel function and had shorter hospital lengths of stay [13].

However, when compared to surgical therapies, percutaneous drainage of liver abscess appears to be associated with lower morbidity and less cost when compared to surgical therapies [5]. In a study of 264 patients with 354 liver abscesses, investigators performed ultrasound-guided aspiration in those with abscesses less than 5 cm and performed ultrasound-guided catheter drainage in those greater than 5 cm. In their series, percutaneous therapy was successful 87 % of the time, with only 8 % converted to operative therapy [14].