Indications for liver transplantation in the setting of PSC include decompensated cirrhosis, recurrent cholangitis, refractory pruritus, and early-stage malignancy not amenable to resection (intrahepatic hepatocellular carcinoma or hilar cholangiocarcinoma) [11]. In many cases, ongoing biliary inflammation, fibrosis, and developing strictures progress to cirrhosis. One-quarter of PSC patients have cirrhosis at time of diagnosis, though less than 4 % present with clinical symptoms of portal hypertension [1, 3]. Presence of ascites or varices at diagnosis is associated with a higher rate of disease progression in the first 5 years [1]. Similar to other forms of chronic liver disease, the presence of complicated cirrhosis warrants evaluation for transplantation.

Recurrent cholangitis is an all too common complication of ongoing PSC and represents an indication for transplantation not found in other chronic liver disease. Again, ongoing biliary inflammation and fibrosis lead to strictures and relative biliary obstruction resulting in recurrent bacterial cholangitis. Frequent hospital admissions and endoscopic interventions lead to a state of chronic illness and poor quality of life. While this morbidity associated with recurrent cholangitis is significant, it does not appear to lead to increased mortality on the waitlist and does not contribute to the patients’ MELD score [12]. This and the scarcity of standard-criteria cadaveric grafts make the practice of petitioning for MELD exception points a continued debate. For this reason, live donor liver transplantation (LDLT) has proven to be a reliable and attractive option for these patients.

Severe, refractory pruritus can be profoundly morbid condition that can cause suicidal ideation and chronic cutaneous excoriations. With these associated conditions, liver transplantation should be considered albeit with the same waitlist and allocation limitations.

The lifetime risk of cholangiocarcinoma (CCA) is 10–15 % in PSC patients, more than one-quarter of which present within 1 year of the initial PSC diagnosis [1–5, 8, 10, 13]. In a select group of patients, CCA is an accepted indication for transplantation with good outcomes. A patient with small, localized, hilar CCA can be transplanted within a rigorous protocol of neoadjuvant chemoradiation and aggressive staging. Tumors that do not fit within this strict protocol are considered contraindications to transplantation with cadaveric grafts.

General criteria for evaluating PSC patients for liver transplantation are similar to criteria for non-PSC patients. This should include a thorough history and physical examination to identify any comorbid medical conditions that are contraindications to transplantation. Although surgical management of extrahepatic bile duct strictures is rare in the setting of PSC, prior biliary surgery for any indication can complicate the liver transplant operation [14]. Cholangiography is the preferred diagnostic intervention for PSC with characteristic findings of segmental bile duct strictures with focal dilations (beading) and mural irregularities of intra- and extrahepatic bile ducts. Liver biopsy may be useful in identifying individuals with small-duct variant of PSC but is not recommended for routine evaluation because PSC does not uniformly affect the liver and a high probability for sampling error exists [15]. Ancillary testing includes laboratory testing, abdominal imaging to evaluate for hepatobiliary masses and assess patency of hepatic vessels, cardiopulmonary testing, bone density assessment, and age-appropriate routine cancer screening (colonoscopy, mammography, Pap smear, PSA level), including tumor markers (CA 19–9 and CEA). Inflammatory bowel disease (IBD) is present in 60–80 % of PSC patients, and PSC patients have a tenfold increased risk of colorectal adenocarcinoma, underscoring the importance of surveillance colonoscopy and IBD management in patients being considered for transplantation [2, 4, 11, 16–19]. A significant subset of PSC patients (approximately 14 %) has other immune-mediated inflammatory or autoimmune disease, which is associated with lower rates of transplant-free survival [20].

Priority on the waitlist for liver transplantation is currently determined by the Model for End-Stage Liver Disease (MELD) score. The MELD score, implemented in 2002 and revised in January 2016, is a formula used to predict 90-day mortality. It is calculated from serum bilirubin, creatinine, INR, and sodium values. As mentioned previously, for PSC patients the MELD system has several limitations. Patients with severe, intractable pruritus or recurrent cholangitis commonly have low INR and creatinine levels, thereby limiting the ability of the MELD score to accurately reflect the severity of disease symptoms and ongoing morbidity [21]. Quality of life can be disproportionately poor in patients with low MELD [22]. MELD also fails to predict the progression of PSC.

Additional MELD points (MELD “exception” points) may be granted to patients with conditions that are not accurately reflected in the MELD calculation, such as a documented history of recurrent cholangitis. It was previously believed that patients with recurrent cholangitis were at increased risk for severe complications, including increased mortality. However, recent analysis of UNOS data suggests that PSC patients with recurrent cholangitis do not suffer higher rates of death and actually have lower waitlist mortality compared to non-PSC patients [12, 21]. Additionally, removal for deterioration in condition is equivalent between PSC and non-PSC groups [21]. In 2006, a consensus conference attempted to narrow upgrade criteria to recurrent bacteremia or sepsis secondary to PSC [23]. Yet, in most regions the petitions continue to be reviewed in a nonstandardized fashion, and a majority of patients are approved with limited symptoms or complications of bacterial cholangitis. Due to minimal comorbid conditions, these patients continue to be transplanted at a high rate [24].

Development of an early-stage primary liver cancer warrants a request for a standardized MELD exception. Liver transplantation offers more than 80 % cure rate for early-stage hepatocellular carcinoma (HCC). Patients with HCC tumors who meet Milan criteria (1 tumor <5 cm or 2–3 tumors each less than 3 cm in diameter) are eligible and have maximal anticipated benefit [25]. As mentioned prior, CCA disproportionately affects patients with PSC, and transplantation for CCA has not yet achieved the high cure rates seen with HCC. Yet, highly specialized treatment protocols such as that published by the Mayo Clinic have improved survival in patients with localized, early-stage, hilar CCA [26]. Pretransplant management includes neoadjuvant chemoradiation, assessment of disease progression (tumor size and vascular invasion) with cross-sectional imaging, lymph node sampling with endoscopic ultrasound, and surgical exploration prior to proceeding with transplantation. The attrition rate is significant, leading to a very selective group of patients, but overall survival is good in multiple high-volume centers [27]. Recurrence of CCA post-liver transplant is high, and only select patients with isolated, hilar CCA, <3 cm, who have completed the protocol with neoadjuvant chemoradiation and aggressive staging should be considered [11, 26].

The waitlist for a graft from a standard donor after brain death (DBD) has become increasingly competitive, with an average MELD score above 30 at time of transplant and wait times greater than 2 years in some regions in the United States [6]. As described above, disease and symptom severity is poorly reflected in the MELD score, and wait times for PSC patients are long. To reduce time to transplantation, PSC patients and transplant surgeons must consider nonstandard donor livers, specifically allografts from donors after cardiac death (DCD), public health service (PHS)-increased risk donors, extended criteria donors (e.g., older donor age), and living donors. The advantages of reducing wait time for transplant in an attempt to reduce waitlist mortality are not without consequence. A retrospective analysis of UNOS data comparing use of DCD vs. DBD livers in PSC patients revealed a significantly increased risk of graft loss in the DCD allograft recipients (hazard ratio 2.4) [28]. The use of DCD livers may disproportionately affect outcomes in PSC patients compared to non-PSC recipients. Specifically, a higher rate of graft loss due to biliary complications has been reported [28]. New protocols for administration of intraoperative hepatic artery thrombolytics with improvement in biliary outcomes in DCD allografts have recently been published, but it remains to be seen if these benefits are realized in the PSC recipient subgroup [29].

PSC patients are more likely to be transplanted with allografts from living donors than are non-PSC patients [30]. Outcomes in LDLT for PSC are better than LDLT done for other chronic liver diseases [31]. When considering live donor options, one must be aware of the potential presence of undiagnosed PSC in family members of recipients. A genetic predisposition exists, with a 100-fold increased risk in first-degree relatives. PSC prevalence is 0.7 % among all first-degree relatives of patients with PSC and 1.5 % among siblings [32]. Certain HLA alleles (B8, DRB1*03, DRB1*13) and other genes have been implicated in the pathogenesis of PSC, but no diagnostic assay exists for which to screen potential donors [33]. Therefore, any family members undergoing donor evaluation should have laboratory testing (including serum bilirubin, AST, ALT, alkaline phosphatase, GGT), cross-sectional imaging (CT or MRI), cholangiography (MR or endoscopic), and possible liver biopsy. A history of inflammatory bowel disease in a potential donor, while not an absolute contraindication, should encourage one to proceed cautiously.

Preparation of the patient in the operating room is similar to that of patients being transplanted for other indications. General anesthesia is administered by an experienced liver transplant anesthesiologist. Adequate venous access for large-volume resuscitation is established, and hemodynamic monitors are placed (e.g., intra-arterial blood pressure catheters, transesophageal echocardiography probe or Swan-Ganz catheter, continuous pulse oximetry). Antibiotic selection and duration in the perioperative period should take into consideration a history of cholangitis in the recipient.

Intraoperatively, many of the technical considerations for the native hepatectomy and vascular reconstruction of the donor liver are the same as for other recipient subgroups. The technical aspect of greatest contention is that of restoration of biliary continuity to the allograft. For most non-PSC recipients, the preferred method is creation of an end-to-end choledochocholedochostomy, or duct-to-duct, anastomosis. Historically, this was not used in PSC patients due to concerns about the risk of residual disease in the extrahepatic bile duct. Reconstruction to the recipient duct was believed to increase the risk of anastomotic stricture and disease recurrence in the allograft. Therefore, Roux-en-Y choledochojejunostomy (RYCJ) was the preferred method for biliary reconstruction in PSC patients, even if there was no gross evidence of disease in the extrahepatic duct at time of surgery. However, RYCJ is not without morbidity. RYCJ configuration can lead to bacterial overgrowth of the biliary system and is significantly associated with an increased risk of ascending cholangitis, 25 % vs. 9 % in duct-to-duct patients within the first year after transplant [34–36]. In some studies, the risk of late development of non-anastomotic strictures is higher in the Roux-en-Y group, which may be related to recurrent inflammation from higher rates of cholangitis [35]. Diagnosis and management of biliary obstruction is more challenging with RYCJ because only the most skilled endoscopists can navigate the Roux limb, thus necessitating percutaneous transhepatic interventions. Anastomotic strictures in RYCJ are more likely to require surgical intervention to correct than strictures with duct-to-duct anatomy [37]. This may be due to difficulty with endoscopic access to Roux limbs. There is also a small, but known, risk of gastrointestinal bleeding from the jejunojejunostomy.