Intra-abdominal and surgical site infections are the most common infection after liver transplant, comprising over half of all infections (Kim et al. 2013) and occurring in 18–51 % of all liver transplant recipients (Freire et al. 2013; Hellinger et al. 2009; Kim et al. 2013; Kim et al. 2008). The infections range from superficial and deep incisional infections to peritonitis, cholangitis, and intra-abdominal abscesses. Posttransplant intra-abdominal and surgical site infections are associated with longer hospital stays and higher medical costs (Hollenbeak et al. 2001) as well as increased death and graft loss (Hellinger et al. 2009).

Risk factors for the development of intra-abdominal and surgical site infections include a higher model for end-stage liver disease (MELD) score at the time of transplant, duration of transplant surgery, Roux-en-Y biliary anastomosis, need for renal replacement therapy following transplant, extended postoperative intensive care unit (ICU) stay, need for reoperation, or extended preoperative hospital stay (Avkan-Oguz et al. 2013).

The biliary tract is a common source for intra-abdominal infections, contributing to the development of cholangitis, abscess, or bilomas. Risk factors for the development of biliary complications in particular, specifically biliary necrosis, strictures, and leaks, include hepatic artery thrombosis, hepatic artery stenosis, Roux-en-Y biliary anastomosis, and T-tube placement (Safdar et al. 2004). As surgical techniques have improved, so too has the incidence of biliary tract complications following liver transplantation. Greif et al. (1994) reported an initial biliary complication rate at their institution of 19 % in 1983 which had fallen to 11.5 % in 1994. Other studies have shown an incidence of 5–25 % (Akamatsu et al. 2011; Chen et al. 2011; Gastaca 2012; Safdar et al. 2004). Of all biliary complications, strictures and leaks are most common, occurring predominantly in the first 3 months after transplant (Akamatsu et al. 2011; Chen et al. 2011; Safdar et al. 2004).

The presentation of intra-abdominal and surgical site infections can range from erythema and pain of the incisional site to fever, abdominal pain, elevated white blood cell (WBC) count and/or liver enzymes, or sepsis . At times, laboratory abnormalities may be the only presenting sign of an intra-abdominal infection. Infections can be caused by a spectrum of bacteria most commonly including Staphylococcus aureus (both methicillin sensitive and methicillin resistant), enterococci, anaerobic bacteria, Escherichia coli, Pseudomonas sp., Enterobacter sp., Klebsiella sp., and Acinetobacter sp. Most infections are managed by surgical or interventional procedures as needed and drainage if a fluid collection is present. Broad-spectrum antibiotics should be initiated immediately and tailored once microbiologic data is available. Treatment should extend for at least 2 weeks once any necessary drainage has occurred.

Bloodstream infections are the second most common infection following liver transplant, occurring in 17–30 % of liver transplant recipients, with mortality reaching 28–36 % (Chen et al. 2011; Kalpoe et al. 2012; Kim et al. 2013; Lee et al. 2011; Singh et al. 2000, 2004). Most bloodstream infections occur within the first 60 days following transplant, although they can occur at any time (Dganga et al. 2012). There are conflicting reports of the predominant bacteria causing bloodstream infections following liver transplant, likely due to center-specific epidemiology, with some documenting primarily gram-negative bacteria (Dganga et al. 2012; Kim et al. 2013; Singh et al. 2004) and others reporting a predominance of gram-positive bacteria (Lee et al. 2011). The timing of bacteremia following liver transplant may play a role, however, as shown by Lee et al. (2011). They found over 92 % of gram-positive bloodstream infections occurred within the first 30 days following transplant, whereas over 41 % of gram-negative bloodstream infections occurred after the first 30 days. For episodes of bacteremia occurring within the first 30 days after transplant, nosocomial bacteria were more common than in episodes of bacteremia occurring after 30 days. Polymicrobial bloodstream infections may also be common, occurring in up to 28 % of liver transplant recipients, primarily comprised of staphylococci, enterococci, and Candida (Kim et al. 2013).

Patients with bloodstream infections may present with a variety of signs and symptoms including fever, rigors, elevated WBC count, sepsis , or localizing symptoms of the primary source. Risk factors for developing bloodstream infections include extended postoperative ICU stay, an MELD score of greater than 20 at the time of transplant, preoperative albumin level of less than 2.8 g/dL, and the need for reoperation (Avkan-Oguz et al. 2013; Singh et al. 2000).

The primary sources of bloodstream infections include intra-abdominal or biliary tract infections, pneumonia, urinary tract infection, intravascular catheter-related infection, or wound infection. Kim et al. (2013) found the primary sources for bacteremia include the biliary tract in 36 %, abdominal or wound in 28 %, and intravascular catheters in 19 % of liver transplant recipients. In patients who developed biliary complications, 42 % developed a concurrent bloodstream infection (OR 2.91 (95 % CI)). Bloodstream infections were found to be an independent risk factor for death with 1 year survival being 60 % in patients who developed a bloodstream infection and 90 % in those who did not (HR 3.93 (95 % CI)). Among patients who developed a bloodstream infection, the risk factors for death included hepatocellular carcinoma (HCC) (HR 3.82), candidemia (HR 3.71), polymicrobial bacteremia (HR 3.18), and posttransplant need for renal replacement therapy (HR 2.44).

Treatment for bloodstream infections involves identification and management of the primary source if possible, removing any indwelling catheters that may be involved and providing broad-spectrum antibiotic coverage that may be tailored once microbiologic data is available. Initial antimicrobial coverage should take into consideration the prevalence of MDR bacteria at each individual institution.