EASL Clinical Practice Guidelines

Management of extrahepatic manifestations


Pruritus can be a feature of any cholestatic disease and can be of sufficient severity, in some instances, also disabling. The precise mechanism of cholestatic pruritus remains unclear [214]. Fluctuation is characteristic (both within the day and over longer periods of time), and pruritus can lessen as end-stage liver disease develops. In the absence of obstructive bile duct lesions amenable to endoscopic, radiological or surgical correction treatment (Fig. 2) focuses entirely on systemic medication (no topical agents have demonstrated efficacy). There is no evidence to suggest that UDCA lessens cholestatic itch (indeed paradoxical worsening of itch has been reported anecdotally following introduction of this agent) except in the context of intrahepatic cholestasis of pregnancy. Cholestyramine is widely used as first-line treatment although the evidence basis to support this is limited, largely because the agent entered widespread use before the era of evidence-based medicine [215]. Poor tolerance due to the taste of this agent can be a problem (which can sometimes be addressed by flavoring with fruit juice). When both agents are used UDCA and cholestyramine should be spaced a minimum of four hours apart to prevent binding and loss of efficacy [216].

Fig. 2 Management of pruritus in cholestasis. Abbreviations: US, ultrasound; MRCP, magnetic resonance cholangiopancreatography; ERCP, endoscopic retrograde cholangiopancreatography.
View Large Image | View Hi-Res Image | Download PowerPoint Slide

The pregnane X receptor (PXR) agonist, rifampicin, is widely used as second-line treatment and has a strong evidence base [[217], [218]]. Ongoing efficacy is reported over up to 2 years of treatment (mirroring clinical experience) [219]. Urine, tears and other body secretions are discoloured during treatment and, in case series, drug-induced hepatitis and significant liver dysfunction after two to three months have been reported in up to 12% of cholestatic patients [220]. In light of this, low dose initiation with monitoring before dose escalation is recommended.

Oral opiate antagonists can be used as third-line agents [218]. However problems have been reported with an opiate withdrawal-like reaction on initiation (which can be ameliorated, to some extent by use of an i.v. naloxone induction phase in which the dose is rapidly escalated to a level at which conversion to the lowest dose oral opiate antagonist preparation can be instituted [[221], [222]] and ongoing problems resulting from pain and confusion.

There is evidence to support the use of sertraline, although the mechanism of its action remains unclear [223]. Clinical experience of both opiate-antagonists and sertraline used for pruritus treatment has been disappointing for many clinicians and the importance of fully exploring the use of cholestyramine and rifampicin therapy before resorting to these agents is emphasized. There are anecdotal observations to support the use of gabapentin and cimetidine in cases of resistance pruritus. The use of antihistamines, ondansetron and phenobarbitone is not recommended for reasons of lack of efficacy, limited efficacy and excessive side-effect profile, respectively.

There is case report evidence to advocate the use of invasive physical approaches in resistant pruritus cases. These approaches include extracorporeal albumin dialysis [224], plasmapheresis [[225], [226]] and bile duct drainage [[176], [227]]. The invasive nature of these approaches makes them only suitable in patients who are resistant to medical therapies. Transplantation is effective for the control of cholestatic itch but raises issues of organ allocation priority and patient risk in patients who would not otherwise require transplantation [228]. Itch quantification using a visual analogue scale can help in the assessment of response to interventions. Objectification of itch through physical measurement of scratching activity has been advocated as a more accurate measure. It is, in practice, limited to use as a research tool. Treatment of pruritus in cholestatic liver disease has been subjected to systematic review [[217], [218]].

Recommendations (Fig. 2)

  1. Cholestyramine 4 g up to four times daily or other resins are regarded as first-line treatment of pruritus (II-2/B1). Resins should be spaced away from UDCA and other drugs by at least 4 hours (II-3/B1).
  2. Rifampicin is regarded as second-line treatment introduced at 150 mg with monitoring of serum liver tests which may be increased to a maximum of 600 mg daily (I/A1).
  3. Naltrexone, an oral opiate antagonist, at a dose of 50 mg daily should be considered as third-line treatment starting at a low dose of 25 mg (I/B1). It should only be considered following proven lack of efficacy, intolerance or side-effects with cholestyramine or other resins and rifampicin (III/C1).
  4. Sertraline may be considered for patients resistant to above mentioned treatments as fourth-line treatment (II-2/C2).
  5. Patients resistant to the above agents can be treated with drugs with anecdotal support, or referred to specialized centers, where more invasive approaches should be considered (III/C2).
  6. Liver transplantation is effective, but should only be considered when all available interventions above have proven ineffective (III/C1).


PBC can be characterized by fatigue, the degree of which is unrelated to the severity of the underlying liver disease. The issue of the extent to which other cholestatic liver diseases can be associated with fatigue is poorly studied. Before ascribing fatigue to PBC it is essential to exclude other associated or non-associated causes of fatigue which may be amenable to specific intervention. This includes the presence of AIH-like features which may be amenable to immunosuppressive therapy. Fatigue in PBC shows only a limited association with depression [229], but stronger associations with autonomic dysfunction (in particular orthostatic hypotension [230]) and sleep disturbance (in particular excessive daytime somnolence [230]) and which may themselves be amenable to specific intervention (there is, in particular, case series evidence to support the use of modafinil in patients with fatigue associated with prominent daytime somnolence [[231], [232], [233]]). There are no specific interventions able to reverse fatigue in PBC, although supportive and understanding clinical care will improve patients' capacity to cope [234]. Fatigue may not improve significantly following liver transplantation which is therefore not appropriate in patients lacking other indications.


  1. Associated disease (e.g., hypothyroidism, anemia, diabetes, depression etc.) or medication use characterized by fatigue should be actively excluded (III/C2).
  2. Supportive measures including minimization of factors likely to exacerbate autonomic dysfunction (e.g., excessive anti-hypertensive medication) and sleep disturbance (e.g., caffeine in the evenings) should be considered (III/C2). Psychological support should be considered to assist with development of coping strategies (II-2 & II-3/C2).
  3. Liver transplantation is not appropriate for treatment of fatigue in the absence of other indications (III/C1).


The degree to which patients with cholestatic liver disease are at increased risk of osteoporosis is unclear, with contradictory reports in the literature. This largely reflects the case mix in different centres (with significant age, disease severity and degree of cholestasis differences). A consensus view would be that patients with end-stage liver disease and/or a high degree of cholestasis are at increased risk of developing osteoporosis, with a significantly smaller risk in other groups. In these latter groups established population risk factors for osteoporosis (smoking, inactivity, family history, low body weight, age and female gender) outweigh any cholestasis-related risk. Compared to healthy controls, male patients with cholestatic liver disease have a higher disease-related osteoporosis risk increase (although lower absolute risk) than female patients. The use of calcium and vitamin D supplements is supported by epidemiological data (reduction or reversal of the natural rate of bone loss) but there are no trial data to support or refute this treatment approach [235]. Hormone replacement therapy is effective in post-menopausal female patients [[236], [237]]. Testosterone therapy should be avoided in male patients because of risk of hepatocellular carcinoma. There are trial data to support the use of bisphosphonates (particularly alendronate) where osteoporosis is present [[238], [239]]. There are limited data to support the use of raloxifene and sodium fluoride [[240], [241]]. Bone mineral density assessment (DEXA) is a useful guide to treatment and should be undertaken where possible in all patients at presentation, with follow-up assessment at between 1 and 5 years depending on outcome and general osteoporosis risk [242].


  1. The risk for osteoporosis should be clinically assessed for all cholestatic patients with emphasis on reversible risk factors and lifestyle advice (III/C2).
  2. Bone mineral density should be assessed by DEXA in chronic cholestatic liver disease at presentation (III/C2). Rescreening should be performed up to annually depending on degree of cholestasis or other individual risk factors (III/C2).
  3. Supplementation with calcium (1000–1200 mg/day) and vitamin D (400–800 IU/day) in all patients with cholestatic liver disease should be considered but is not evidence-based (III/C2).
  4. Alendronate or other bisphosphonates are indicated at a T score <−2,5 (DEXA) or following pathological fracture (I/B1) and may be appropriate at a T score <−1,5 (III/C2).

Fat-soluble vitamin substitution

Fat malabsorption can complicate highly cholestatic disease variants, although the risk is lower in less cholestatic patients than has previously been considered to be the case (with the exception of children where degrees of fat malabsorption are typically higher). Parenteral vitamin K supplementation should be given prophylactically in overt cholestasis prior to any invasive procedure and in the context of bleeding Assessment of blood levels of fat-soluble vitamins has been advocated to guide the need for supplementation but this approach is not widely used and is not recommended.


  1. Calcium and vitamin D enteral supplementation should be considered in all cholestatic patients as part of the osteoporosis prevention protocol (III/C2).
  2. Vitamin A, E and K should be supplemented enterally in adults in the context of overt cholestasis, where the clinical features of steatorrhea are present or where fat-soluble vitamin levels are proven to be low (III/C2).
  3. Parenteral vitamin K should be given prophylactically prior to invasive procedures in overt cholestasis and in the context of bleeding (II-2/C1).

Varices and hepatocellular carcinoma

Varices and hepatocellular carcinoma (HCC) development occur in advanced cholestatic liver disease as in other forms of chronic liver disease and are associated with impaired prognosis [[243], [244]]. Screening, prophylaxis and treatment approaches should be adopted as in other chronic liver disease settings [[245], [246]]. However, a platelet count of <200,000/mm3, serum albumin <40 g/L and serum bilirubin >20 μmol/L were independent risk factors for the presence of oesophageal varices in one cohort of cholestatic patients with PBC (>90%) and PSC [247]. The proposed threshold of endoscopic screening for oesophageal varices may be valid for PBC rather than cholestatic liver disease in general.