Ursodeoxycholic acid (UDCA) is a hydrophilic, tertiary bile acid which has been used for the treatment of a variety of chronic cholestatic conditions [59]. It has been shown to be an effective therapy in PBC [60]. After oral administration, UDCA is absorbed mainly in the small intestine, and then it has an enterohepatic circulation. At a daily dose of 13–15 mg/kg body weight/day, UDCA constitutes 40–50 % of total bile acid pool and results in a decrease in relative contribution of the more hepatotoxic endogenous hydrophobic bile acids [59]. Several mechanisms have been proposed by which UDCA may protect against cholestatic liver injury, and these include the choleretic effect by increasing bile flow, protection of injured cholangiocytes from toxic bile acids, stimulation of detoxification of hydrophobic bile acids, inhibition of hepatocyte apoptosis, and direct cytoprotective and immunomodulatory effects [29, 59].

The majority of the early studies of UDCA in PSC were small and/or uncontrolled. Many of these studies demonstrated liver function test improvement by using doses of 10–15 mg/kg body weight/day [8, 26, 61]. Lindor et al. conducted a randomized controlled trial (RCT) of UDCA 13–15 mg/kg/day, for 2–5 years, in 105 PSC patients. The results demonstrated improvement in LFTs but not symptoms and the time to treatment failure (defined by histologic progression by two stages, development of cirrhosis or esophageal varices, liver decompensation, LT, or death) [62]. On the basis that higher doses of UDCA may be required to provide sufficient delivery of UDCA to the bile pool and also enhance immunomodulatory effects in the setting of cholestasis and bile duct injury in PSC, several studies using higher dose of UDCA were conducted and published in the early 2000s. An RCT from Oxford using UDCA 20–25 mg/kg/day in 26 PSC patients found significant improvement in LFTs, histology, as well as cholangiographic features. However, no benefit in symptoms and survival was demonstrated [63]. Two studies comparing different doses of UDCA suggested that higher daily dose (25–30 mg/kg) was well tolerated and provided benefits, which included survival benefit in one study, compared to a lower dose (10–20 mg/kg) [64, 65].

Despite somewhat convincing data on benefits with higher doses, a large Scandinavian RCT evaluating UDCA at 17–23 mg/kg/day in 219 PSC patients for 5 years found no significant favorable effect on survival, symptoms, and prevention of CHCA although there was a nonsignificant trend toward improvement in LFTs and survival [66]. Recently, a multicenter RCT from the United States comparing high-dose UDCA (28–30 mg/kg/day) with placebo, in 150 PSC patients, was terminated prematurely at 6 years due to a higher incidence of adverse outcomes (i.e., death, LT, esophageal varices) in the UDCA group [67]. The likelihood of developing serious adverse outcomes was not predicted by biochemical response, but was predicted by advanced liver disease at presentation [67]. Therefore, currently there is no established role for UDCA in slowing the progression of PSC. Further, high-dose UDCA may be harmful in late-stage disease [8, 60].

Unlike most of other immune-mediated diseases, treatment with corticosteroids and other ISAs has not demonstrated consistent benefits in PSC, and most evidence is derived from pilot studies. Corticosteroids demonstrated no benefit in PSC and were associated with significant worsening of osteoporosis [68–70]. Corticosteroids may be considered only in patients with PSC/AIH overlap and IgG4-associated cholangitis [8]. No controlled trial of azathioprine (AZA) has been reported as monotherapy to date. A combination of AZA, prednisolone, and UDCA (500–750 mg/day) for PSC was reported in a case series of 15 PSC patients. All patients had ALP improvement (seven patients had been previously treated with UDCA, but ALP improved only after prednisolone and AZA were added), and 60 % had histological improvement after 41 months [71]. Methotrexate (MTX) may minimally improve ALP levels, but does not impact clinical outcomes of PSC [72]. Addition of MTX to UDCA was associated with toxicity and without further improvement in LFTs [73]. Mycophenolate mofetil (MMF) was poorly tolerated (56 %) and did not demonstrate clinical benefit in PSC [74]. Further, combination of MMF and UDCA did not provide additional benefits [73]. Tacrolimus [75] and cyclosporin [76] had no significant effects on PSC disease outcomes and were poorly tolerated although they provided benefits in UC [76]. A pilot RCT in 10 PSC patients failed to demonstrate clinical efficacy of infliximab (5 mg/kg) on the course of liver disease [77].

Based on the observation that elevated hepatic copper levels are universally detected in patients with chronic cholestasis, D-penicillamine was evaluated in an RCT of 70 PSC patients. However, it was not associated with clinical benefit and has considerable toxicity, which led to treatment discontinuation in 21 % of patients [78]. Colchicine, an anti-fibrogenic agent, either alone or in combination with prednisone failed to show beneficial effects in two RCTs [70, 79]. Silymarin, a milk thistle extract, which potentially has several hepatoprotective properties, was evaluated in an open-label pilot study of 30 PSC patients for 1 year [80]. One-third of patients achieved substantial improvement in LFTs, but no significant change in Mayo risk score [80]. Several agents, such as nicotine, bezafibrate, pirfenidone, minocycline, and probiotics, have been preliminarily evaluated in PSC but without clear demonstrable benefits [8, 61, 81].

CHCA, a dreadful complication of PSC, develops in 8–15 % of patients [5, 7, 9]. Risk factors include the duration of UC, colonic dysplasia, variceal bleeding, proctocolectomy, alcohol consumption, and polymorphisms in the NKG2D gene [8, 82]. The diagnosis of CHCA in the setting of PSC is often challenging, particularly for the periductal infiltrative type. The presence of mass-like lesion or a long biliary stricture, particularly in the hilar area, on an imaging study strongly raises the possibility of CHCA. In PSC patients with clinical suspicion for CHCA, CA 19-9 at a cutoff of 129 U/mL has value in determining the likelihood for CHCA; positive predictive value was 57 % and negative predictive value 99 % [83]. However, caution must be exercised since CA 19-9 is undetectable in person with Lewis-negative blood type and can be elevated in other conditions (i.e., cholangitis, non-biliary cancers) [84]. A combination of cross-sectional liver imaging studies, tumor biomarkers, and cholangiography with tissue sampling is often required, and is recommended, to make the accurate diagnosis of CHCA (Fig. 41.3) [8, 54, 84].

The prognosis of CHCA in PSC is dismal with a 3-year survival of less than 20 % even in surgically resected patients [8]. Recent data suggests that neoadjuvant therapy followed by LT in highly selected patients may result in a better outcome with a 5-year survival of ~70 % [85]. The benefit of other palliative modalities, such as external beam radiation, endoscopic ablative therapy, and systemic chemotherapy, has not been clearly demonstrated [8].

PSC has been shown to be an independent risk factor for the development of colorectal neoplasia in patients with UC (OR 4.79, 95%CI; 3.58–6.41) [86]. Colorectal neoplasia associated with PSC can be diagnosed at any time during the course of PSC, and it appears to occur predominantly in the right colon [87]. This risk appears to persist even after LT [5, 24]. Therefore, colonoscopy surveillance for colonic neoplasia is recommended to begin at the time of the diagnosis of PSC [8]. There are controversial data suggesting a benefit of UDCA in preventing the development of colorectal neoplasia [8, 60]. The current US guideline recommends against the use of UDCA as chemoprevention in patients with PSC-UC [8], while the European guideline suggests the use of UDCA in high-risk patients, such as those with strong family history of colorectal cancer, previous history of colorectal neoplasia, or long-standing extensive colitis [60].

Gallbladder abnormalities are commonly observed in patients with PSC, and these include gall stones (26 %), PSC involving the gallbladder (15 %), and gallbladder neoplasms (4–14 %) [8, 88]. Remarkably, 40–60 % of gallbladder polyps detected in patients with PSC are malignant [89, 90]. Therefore, surveillance by ultrasound should be done annually. In patients with a gallbladder mass lesion, cholecystectomy is recommended regardless of lesion size since the 1-cm rule may not reliably predict malignant potential of the gallbladder polyp in the setting of PSC [8, 91].

Patients with PSC are susceptible to repeated episodes of bacterial cholangitis, especially after biliary tract manipulation [92]. If cholangitis occurs without biliary intervention, the presence of stones, dominant strictures, or CHCA should be considered. Most common causative organisms are gram-negative enteric bacteria and enterococci [92]. The majority of patients respond to broad-spectrum intravenous antibiotic plus biliary drainage. Patients with recurrent bacterial cholangitis may benefit from long-term antibiotic prophylaxis [8].

Management of portal hypertension and its complications in patients with PSC does not differ from other etiologies. The ultimate treatment for end-stage liver disease (ESLD) associated with PSC is LT with 5-year survival rates of approximately 85 % [8]. Resection of the extrahepatic biliary tree along with a Roux-en-Y choledochojejunostomy is widely accepted as a method of choice for biliary reconstruction in LT for PSC [60]. As in non-PSC, the Model for End-Stage Liver Disease (MELD) score is most widely utilized for organ allocation for PSC patients, although the presence of dominant strictures may affect MELD score by increasing bilirubin levels, and this may not necessarily mean advanced disease and liver failure. Other unique indications for LT in PSC patients include intractable pruritus, recurrent bacterial cholangitis, and CHCA [8]. Recurrence of PSC occurs in 20–25 % of the liver grafts after 5–10 years following LT [8, 93], but this is sometimes difficult to assess due to the similarities in biliary changes seen with ischemic and preservative injury, infections, and chronic rejection.

The activity of UC following LT is heterogeneous. Contrary to general wisdom, while on liver transplant-related immunosuppression, the majority of PSC-IBD patients experience a deterioration of their IBD following LT [94]. Further, the increased risk of developing colorectal neoplasia persists after LT [95]. The guideline for the management of UC exacerbation after LT has not been established, and the long-term effect of anti-TNF agents on liver graft is unknown.

Patients with long-standing IBD, and particularly with the prolonged use of corticosteroid therapy, frequently have decreased bone mass density (BMD) [96]. The presence of PSC, with or without cirrhosis, further negatively impacts BMD by several mechanisms, such as vitamin D malabsorption, altered bone turnover rate, and hypogonadism [97]. A recent study of 237 PSC patients with 10 years of follow-up reported that patients with PSC lost 1 % of their BMD per year. Osteoporosis was detected in 15 % of patients (24-fold higher rate than matched population) and risk factors included older age, low body mass index, and long duration of IBD [98]. The surveillance and management of osteoporosis in PSC does not substantially differ from other situations, and there is particular emphasis on calcium and vitamin D supplementation [8, 96]. Oral bisphosphonates may induce esophageal ulcerations which could precipitate variceal hemorrhage. Therefore, parenteral bisphosphonates may be a reasonable approach for patients with esophageal varices [8].

Small-duct PSC, previously termed as pericholangitis, refers to a subgroup of patients who have biochemical and histological features compatible with PSC, but with normal cholangiography. Small-duct PSC represents approximately 6–11 % of patients with sclerosing cholangitis and often coexists with IBD (~80 %). It is a disease that is potentially progressive but is associated with a better long-term prognosis as compared with large-duct PSC (LT-free survival 13 years vs. 10 years, respectively; p < 0.0001) [99]. Cholangiocarcinoma does not seem to occur in patients with small-duct PSC. Approximately one-fourth of patients eventually progressed to large-duct PSC over a median of 7.4 years, and some patients progressed to end-stage liver disease requiring LT without developing large-duct disease [99]. Given a relatively small number of patients with small-duct PSC, the management is not well defined, and there is no controlled prospective study reported to date. In a longitudinal cohort of 42 patients from Mayo Clinic followed up to 25 years, UDCA 13–15 mg/kg/day improved liver biochemistries, but did not significantly delay disease progression [100].

PSC/AIH overlap is an ill-defined immune-mediated disorder, which is predominantly encountered in children and young adults [101]. A diagnosis of PSC/AIH overlap is made when both typical cholangiographic features of PSC and definitive diagnosis of AIH, based on modified AIH score, are present [101, 102]. The prevalence of PSC/AIH overlap in patients with PSC has varied from 7 to 14 % based on the revised AIH criteria, and majority of these patients (50–88 %) have underlying IBD [101]. The presentation of PSC/AIH overlap may be either simultaneous or sequential. Particularly in the setting of IBD, patients with PSC with an elevation of ALT should prompt a search for AIH. On the other hand, PSC should be considered in AIH patients with cholestasis, histological bile duct injury, and in those who show a poor response to therapy [101]. Patients with PSC/AIH overlap seem to benefit from UDCA and ISA, and survival is apparently better than in classical PSC, but with a poorer outcome than AIH [103, 105]. In a prospective Italian study (N = 7 PSC/AIH, 34 PSC), a combination of UDCA (15–20 mg/kg/day), prednisolone (0.5 mg/kg/day, then tapered to 10–15 mg/day), and azathioprine (50–75 mg) reported a good biochemical response (ALT, but not ALP) in patients with PSC/AIH overlap [104].

IgG4-associated cholangitis (IAC) is a biliary tract disease of unclear pathogenesis and with cholangiographic features indistinguishable from PSC. It has been described in patients with autoimmune pancreatitis (AIP) as a part of a systemic autoimmune process associated with IgG4 [61, 105]. The clinical entity is characterized by pancreatic enlargement, elevated serum IgG4 levels, histologic evidence of lymphoplasmacytic infiltrate in the pancreas, and extrapancreatic manifestations, such as sclerosing cholangitis, sialadenitis, and retroperitoneal fibrosis [61, 105]. It is important to distinguish PSC from IAC, which is at times challenging due to the fact that pancreatic disease may not be evident (8 %) and IgG4-associated sclerosing colitis, mimicking IBD or UC itself, may be present [61, 105, 106]. IAC appears to be histologically distinct from PSC, and it usually has a dramatic response to corticosteroids in contrast to the refractory nature in PSC [5, 61]. All patients with possible PSC should be tested for serum IgG4 levels to exclude IAC [8]. Interestingly, up to one-quarter of patients with biopsy-proven PSC have an increased IgG4 periductal plasma cell infiltrate, and 9–22 % have elevated serum IgG4 levels [107, 108]. These findings raise the question of whether PSC and AIP/AIC represent different ends of the same disease spectrum or are distinct disease entities, although current evidence favors the latter [8]. Of note is that PSC patients with elevated IgG4 levels tend to have more severe disease severity and course [107, 108]. Thus, a trial of corticosteroids may be considered; however, proven benefit has yet to be demonstrated in RCT [61, 108].

Nonalcoholic fatty liver disease (NAFLD) is the most common hepatobiliary disease encountered in IBD, with reported prevalence range of 2–80 % (median 29.5 %) in published studies with over 100 subjects [109]. Thus, it may account for approximately 40 % of cases of LFT abnormalities in those with IBD [3]. As in the general population, the prevalence of NAFLD is likely increasing among IBD population as well [109]. Apart from obesity and metabolic syndrome, NAFLD can be caused by IBD-related factors, such as malnutrition, protein loss, and medications (i.e., corticosteroids, MTX, anti-TNFs) [109]. There is no specific guideline for the management of NAFLD in IBD patients; periodic monitoring and counseling to avoid excessive weight gain would be reasonable.

Portal vein thrombosis (PVT) is rarely observed in IBD patients in a nonsurgical setting, but it is not uncommon in those with recent abdominal surgery [5]. Several factors associated with IBD, such as increased platelet counts, factor V, VII, and fibrinogen levels, state of active bowel inflammation, and infections, may contribute to the thrombotic complications [5]. Portal vein thrombosis has been detected by CT imaging in 25–45 % of UC patients who have undergone IPAA surgery and was more likely to be segmental, multiple, and occlusive [110–112]. Presentation of PVT following IPAA surgery includes abdominal pain, nausea/vomiting, prolonged ileus, and leukocytosis. Septic complications (i.e., liver abscesses, pelvic sepsis) subsequently occur in 0–50 % of patients [111, 112]. Anticoagulation and antibiotic treatments are generally associated with a good clinical response and resolution of PVT. Interestingly, postoperative PVT has been linked to an increased risk of subsequent pouchitis (46 % in patients with PVT and 15 % in those without after 36 months follow-up) [111].

Several other hepatobiliary diseases have been reported to be more prevalent in IBD patients than in the general population, and these include PBC, gallstones, hepatic amyloidosis, and granulomatous hepatitis [5] (Table 41.2). These conditions have been reported more often in CD than UC (with the exception of PBC), and the mechanisms are unclear. UC patients with PBC typically present with elevated ALP and positive AMA. Interestingly, unlike classical PBC, UC-associated PBC seems to occur at a younger age and more often in men [113]. Hepatic amyloidosis classically occurs in patients with long-standing active IBD and presents with elevated ALP and hepatomegaly (with hard consistency). Involvement of other organs, particularly gastrointestinal tract and kidneys, is commonly observed [114]. Granulomatous hepatitis can be linked to either IBD itself (CD) or IBD-related factors, such as medications, particularly sulfasalazine, and infections. It often presents with isolated elevation of ALP, although other features, such as fever and hepatomegaly, can also be encountered [5, 115]. The diagnosis of hepatic amyloidosis and granulomas generally requires a liver biopsy.