Problems in Ulcerative Colitis That Won’t Go Away Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
Primary Sclerosing Cholangitis Kate D. Williamson a, b Roger W. Chapman a, b a
Nuffield Department of Medicine, Oxford University, and b Translational Gastroenterology Unit, John Radcliffe Hospital, Oxford, UK
Abstract Primary sclerosing cholangitis (PSC) is a chronic, cholestatic liver disease caused by diffuse inflammation and fibrosis that can involve the entire biliary tree. It is a progressive disorder which can ultimately lead to biliary cirrhosis, portal hypertension and hepatic failure. PSC is a complex genetic disorder with male predominance. Environmental predisposing factors include non-smoking. It is closely associated with inflammatory bowel disease (IBD), particularly ulcerative colitis, which occurs in about two thirds of PSC cases. Recent studies have suggested that PSC-IBD is a separate disease entity from IBD alone with distinctive genetic and phenotypic characteristics. Most PSC patients are asymptomatic at presentation; clinical symptoms include fatigue, jaundice, weight loss, right upper quadrant pain and pruritis. Serum biochemical tests indicate cholestasis, and diagnosis is usually established by cholangiography. In symptomatic patients, median survival from presentation to death or liver transplantation is about 12 years. It is a premalignant condition, and the majority of deaths are from malignancy, particularly cholangiocarcinoma or colonic cancer. PSC has no curative treatment. Medical treatment with ursodeoxycholic acid may slow progression of the disease and reduce colon-
© 2014 S. Karger AG, Basel 0257–2753/14/0324–0438$39.50/0 E-Mail [email protected] www.karger.com/ddi
ic dysplasia, though trials lack statistical significance. Liver transplantation is the only option in young patients with PSC and advanced liver disease. © 2014 S. Karger AG, Basel
Introduction
Primary sclerosing cholangitis (PSC) is a progressive cholestatic liver disease characterised by inflammation and fibrosis of both small and large bile ducts of the liver. Although sclerosing cholangitis was first described in the 1860s by Hoffman [1], only a few cases were described over the next 100 years, usually at post-mortem. However, the advent of endoscopic retrograde cholangiopancreatography (ERCP) in the 1970s enabled the diagnosis of PSC to be made ante mortem. Diagnosis was further improved by magnetic resonance cholangiopancreatography (MRCP) in the 1990s. The clinical picture can range from an incidental finding of a raised alkaline phosphatase (ALP) in an asymptomatic patient, to fatigue, lethargy and pruritis. It may progress to cirrhosis of the liver and subsequent liver failure leading to death or liver transplantation. Although PSC is considered to be a rare disease, its incidence is increasing around the world. The exact aetiopathogenesis remains unknown, but it is immune mediated, occurring in genetically predisposed individuals. There is a strong Roger W. Chapman Translational Gastroenterology Unit, Level 5 John Radcliffe Hospital, Headley Way, Headington Oxford OX3 9DU (UK) E-Mail Roger.chapman @ ndm.ox.ac.uk
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
Key Words Cholangiocarcinoma · Cholestasis · Inflammatory bowel disease · Primary sclerosing cholangitis
Epidemiology
PSC is a rare disease, which is predominantly manifested in men, with a male:female ratio of around 2: 1. Whilst the disease can present at any age, the mean age of presentation is 40 years. Prevalence rates are of the order of 0.22 per 100,000 to as high as 16.2 per 100,000 [2–4]. A recent systematic review found the incidence to be as high as 1.3 per 100,000 per year in some populations [5]. The incidence and prevalence of PSC varies according to geographic distribution, with an increasing prevalence in Northern Europe. The incidence appears to be increasing worldwide, though it is not clear whether this is a true increase, or reflects improved detection and diagnosis [6]. There is an inherited genetic predisposition to the disease, which is discussed further below, with first-degree relatives of affected patients having a 10- to 20-fold increased risk of developing the disease above that of the general population. Environmental factors are also involved with the absence of cigarette smoking appearing to be a predisposing risk factor [7, 8], which is independent of the presence of IBD. In one study, only 4.9% of patients with PSC smoked compared with 26.1% of controls [9]. There is a strong association with IBD which is dependent on geographic location. About two thirds of Northern European patients with PSC will have IBD, compared with about half of Southern Europeans, and only 20% of Japanese patients.
Aetiology
The exact aetiopathogenesis remains unknown, but it is established as being immune mediated, occurring in genetically predisposed individuals and strongly associated with IBD. The reason(s) for the close association of PSC with IBD is unclear. Theories include exposure of the liver and biliary system to antigens derived from bacteria present in the ‘biome’ through increased bowel wall permeability – the so-called ‘leaky gut’ theory. Inappropriate homing of gut-derived memory T cells to the liver has Primary Sclerosing Cholangitis
also been suggested as playing a role. These hypotheses are not mutually exclusive. Genetic Factors Many studies have investigated the genes associated with PSC. Initially, candidate gene studies were employed, followed by genome-wide association studies from Northern Europe [10]. Most recently, an international multicentre immunochip-based study was published, comparing 130,422 single nuclear polymorphisms in 3,789 patients with PSC with 25,079 controls [11]. This large study confirmed previous studies that PSC is immune related, and strongly associated to specific human leucocyte antigens (HLA) encoded on chromosome 6. Many HLA associations and other immune response genes were confirmed in the immunochip project to be common to and shared with other known autoimmune diseases, such as coeliac disease, IBD, type 1 diabetes mellitus and multiple sclerosis. Many studies have established that HLA-A1 B8 DR3, HLA-D2 and HLA-DR6 are the three genes most strongly associated with PSC, and these are evenly distributed amongst patients with PSC, regardless of whether they have IBD. It has been suggested that DR3, DR6 and DR2 encode for amino acids in the HLA β-chain that may enhance antigen presentation by the HLA molecule to the T-cell receptor. The FUT2 gene has recently been associated with PSC and Crohn’s disease, and is known to affect the behaviour of the intestinal flora [10]. Other Proposed Mechanisms Other factors are felt to play a role in aetiology, such as a toxic effect of bile on damaged biliary epithelium; the so-called ‘leaky gut’ theory, where pre-existing IBD predisposes to increased bowel wall permeability, and therefore increased exposure of the bile ducts to bacteria and other pathogens; and also an increase in homing T cells and other cytokines from the bowel to the liver, involving upregulaton of VAP-1, CCR9 and various other chemokines [12]. Descriptions of these theories are beyond the scope of this paper.
Clinical Features
Symptoms and Signs The clinical presentation of PSC is variable. Nowadays, most patients are asymptomatic, with investigation prompted by an incidental finding of a raised ALP on a routine blood test, typically in a young to middle-aged Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
439
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
association with inflammatory bowel disease (IBD), which often runs an independent course from the liver disease. It is overrepresented in men compared with women (male:female ratio of 2:1). PSC is also associated with an increased prevalence of hepatobiliary and colorectal malignancy, which accounts for a significant proportion of morbidity and mortality.
Color version available online 50 μm
Fig. 1. MRCP image showing typical features in PSC with strictur-
man with co-existing IBD. Conversely, patients may suffer from lethargy, intermittent jaundice, weight loss, right upper quadrant discomfort and pruritis. Rarely, patients present with decompensated liver disease. During the course of the disease, patients may occasionally develop symptoms of acute cholangitis, such as fever, rigors and abdominal pain, due to infection proximal to a dominant stricture. Physical examination findings are few. Hepatomegaly and splenomegaly are the most common findings in patients with advanced disease. Serology Serum biochemistry typically reveals a cholestatic picture, with ALP usually elevated to at least three times the upper limit of normal and raised γ-glutamyltranspeptidase. A normal ALP does not preclude the diagnosis if suspected. Serum aminotransferases are often mildly raised to 2–3 times the upper limit of normal, though they can be normal. Bilirubin is uncommonly raised at presentation, but often rises in the later course of the illness, representing more advanced disease. Up to 60% of patients have a mildly raised serum IgG level, up to 1.5 times the upper limit of normal, and are almost always raised in children. Around half of patients have raised IgM levels. Serum IgG4 levels are raised in 9–26% of patients with PSC, compared with only 1% of patients with another biliary disease, primary biliary cirrhosis [13–16]. It is important to test serum IgG4 levels in all patients with PSC, as elevated levels (>1.4 g/l) may confer a poorer prognosis [13, 15]. 440
Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
Fig. 2. Part of a portal tract in PSC, showing a bile duct (long arrow) with mild reactive changes in the epithelium and with concentric (‘onion skin’) fibrosis. An adjacent hepatic artery branch (short arrow) is also seen. Only a very light chronic inflammatory cell infiltrate comprising lymphocytes and plasma cells is present. HE. Magnification ×200. Photo courtesy of Adrian Bateman, Southampton, UK.
Antibodies against cystoplasmic antigens in neutrophils (ANCA) are detected in 80% of PSC patients, though it is not specific – ANCA are present in 30–40% of patients with ulcerative colitis (UC) and may potentially serve as a risk factor for developing PSC. The type of ANCA is different from those in vasculitides, which are against the proteinase PR3 and myeloperoxidase. It is unclear whether ANCA have a pathogenic role in PSC, but they may be useful to aid diagnosis.
Diagnosis
The diagnosis of PSC may be made in the setting of chronic cholestasis on biochemistry (elevated ALP and γ-glutamyltranspeptidase), with supportive cholangiographic findings. Usually, MRCP is employed as the diagnostic imaging modality of choice, with ERCP, given its associated risks, reserved only for therapeutic options and/or poor-quality MRCP (fig. 1). If the cholangiogram is normal, but PSC is suspected, a liver biopsy is indicated. Whilst the pathological findings in PSC can be highly variable, characteristic features include portal tract inflammation with lymphocytes, progressing to obliterative concentric fibrosis (so-called ‘onWilliamson /Chapman
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
ing, beading and dilatation of both intra- and extrahepatic ducts. Photo courtesy of Helen K. Bungay, Oxford, UK.
ion skinning’) and bile duct destruction (‘ductopenia’; fig. 2). The basement membrane of the bile duct is often thickened, and copper deposition is usually present. In the setting of cholestasis, a normal cholangiogram and a liver biopsy supportive of PSC, a diagnosis of smallduct PSC may be made. This confers a good prognosis for the patient, with very little chance of progressing to advanced fibrosis and liver decompensation and/or death unless it progresses to classical PSC. There have been no cases of hepatobiliary or colorectal malignancy reported in smallduct PSC. Approximately 20–25% of patients with smallduct PSC progress to classical PSC. A change in the clinical picture, such as the development of pruritis or increases in bilirubin or ALP, for example, should prompt the clinician to re-image the patient with small-duct PSC to look for cholangiographic transition to classical PSC, in which case the prognosis conferred is then that of classical PSC. The other indication for a liver biopsy in PSC is in the setting of suspicion of overlap with autoimmune hepatitis, when there are markedly elevated transaminases and/ or IgG levels, which may occur in up to 10% of cases. Secondary causes of sclerosing cholangitis must be excluded, such as previous biliary surgery, bile duct stones, toxic damage to bile ducts, e.g. intrahepatic injection of formalin, 5-fluorodeoxyuridine and alcohol, AIDS and IgG4-associated cholangitis (for which HISORt criteria may be applied) [17].
without PSC, and IBD often pre-dates the diagnosis of PSC, though not always. In one large study of a well-defined Dutch cohort of 380 patients with PSC and IBD, 75% had UC, 21% had Crohn’s disease, and 4% had IBD-unspecified [23]. Of those with UC, 83% had a pancolitis, 13% a left-sided colitis, and 4% had proctitis only. Of those with Crohn’s disease, 95% had ileocolitis, and only 5% had ileitis alone. In a subgroup analysis of 80 PSC-IBD patients compared with 80 IBD controls, 85% of PSC-IBD patients had inflammation of the right hemicolon, compared with 54% of those with IBD alone. Interestingly, in contrast to PSC-UC, the clinical activity of Crohn’s colitis in patients with PSC is similar to patients who have Crohn’s colitis without PSC [24]. The clinical activity of IBD is not related to the severity of PSC and appears to run an independent course. Having said that, there is evidence suggesting that the more severe the course of PSC, the milder the form of UC, with fewer flares and a reduced need for surgery [25]. Interestingly, when a patient undergoes liver transplantation for PSC, their IBD may worsen in severity, raising the issue of whether the presence of PSC has a protective effect on the course of IBD [26, 27]. However, a recent study contradicts this, displaying a lower index of clinical activity in those with PSC-IBD who have undergone liver transplantation compared with those who have not [28].
Osteoporosis IBD in PSC
Primary Sclerosing Cholangitis
As with other chronic liver diseases, there is an increased prevalence of osteoporosis in PSC, of the order of 15–30% [29]. Its presence does not appear to depend upon the severity of the underlying PSC [30]. European and British guidelines recommend a bone mineral density assessment by DEXA on diagnosis of PSC, as well as consideration towards supplementation with calcium and vitamin D, though this is not evidence based [31, 32].
Natural History of the Disease
Progression of Disease to Liver Failure The natural history of PSC is highly variable among patients. Observational studies have suggested the median time from diagnosis to death or liver transplantation is 10– 12 years, though in asymptomatic patients, 75% of patients survive 15 years or more. A recent population study from the Netherlands gave a much more optimistic view of a median time of 21.3 years from diagnosis to PSC-related Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
441
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
As discussed previously, there is a strong association of PSC with IBD. All patients diagnosed with PSC should be evaluated for IBD with a full colonoscopy if a pre-existing diagnosis of IBD does not yet exist. Around 75– 80% of the concomitant IBD is UC, with 10–15% being Crohn’s disease (predominantly colitis) and 5–10% being IBD-unspecified (colitis). Conversely, around 3–10% of patients with IBD will develop PSC, and persisting elevation in ALP in the patient with IBD (particularly pancolitis) should prompt investigation [18–20]. The phenotype of IBD in the setting of PSC has been suggested to be distinct from IBD without PSC [21, 22]. IBD is predominantly a pancolitis, and it is usually worse in severity on the right compared with the left. Backwash ileitis is more common in UC associated with PSC compared with UC alone, and the IBD is usually relatively less symptomatic than its endoscopic appearance would suggest, often running a fairly indolent course clinically. IBD is diagnosed at an earlier age in PSC, compared with IBD
100
ting of acute cholangitis related to a dominant stricture, and prophylactic antibiotics, such as low-dose ciprofloxacin, are sometimes required in the setting of recurrent cholangitis.
Population-based cohort LT centres cohort
60
Malignancy 40 20 0
Patients at risk
p < 0.001 0 590 422
5 10 15 20 25 30 Time since diagnosis until LT or PSC-related death (years) 378 266
206 143
104 67
50 26
18 9
5 0
Fig. 3. Survival until liver transplantation or PSC-related death of PSC patients in a large population cohort compared with a liver transplantation (LT) centre cohort (figure courtesy of Boonstra et al. [33]).
death or liver transplantation [33]. This may be a more realistic figure, as it represents a large population, as opposed to tertiary centres with potential referral bias (fig. 3). A well-characterised Swedish cohort of 604 patients with PSC [34] showed a mortality rate of 28% during a median follow-up of 5.7 years, with 44% of deaths due to cancer and 37% due to liver disease. 15% of the whole 604-patient cohort had undergone liver transplantation, and 23% of patients experienced cancer at some time (table 1). Levels of ALP fluctuate over the course of the illness, but about a third of patients may have a persistent reduction in ALP below 1.5 times the upper limit of normal. If this occurs, it is felt to be a positive prognostic factor, with only 3% going on to liver decompensation, liver transplantation or liver-related death [35, 36]. Dominant Strictures Patients may develop progressive jaundice, worsening of their liver biochemistry or symptoms of cholangitis, prompting investigation for a dominant stricture. A dominant stricture is defined as a stricture
Primary Sclerosing Cholangitis Kate D. Williamson a, b Roger W. Chapman a, b a
Nuffield Department of Medicine, Oxford University, and b Translational Gastroenterology Unit, John Radcliffe Hospital, Oxford, UK
Abstract Primary sclerosing cholangitis (PSC) is a chronic, cholestatic liver disease caused by diffuse inflammation and fibrosis that can involve the entire biliary tree. It is a progressive disorder which can ultimately lead to biliary cirrhosis, portal hypertension and hepatic failure. PSC is a complex genetic disorder with male predominance. Environmental predisposing factors include non-smoking. It is closely associated with inflammatory bowel disease (IBD), particularly ulcerative colitis, which occurs in about two thirds of PSC cases. Recent studies have suggested that PSC-IBD is a separate disease entity from IBD alone with distinctive genetic and phenotypic characteristics. Most PSC patients are asymptomatic at presentation; clinical symptoms include fatigue, jaundice, weight loss, right upper quadrant pain and pruritis. Serum biochemical tests indicate cholestasis, and diagnosis is usually established by cholangiography. In symptomatic patients, median survival from presentation to death or liver transplantation is about 12 years. It is a premalignant condition, and the majority of deaths are from malignancy, particularly cholangiocarcinoma or colonic cancer. PSC has no curative treatment. Medical treatment with ursodeoxycholic acid may slow progression of the disease and reduce colon-
© 2014 S. Karger AG, Basel 0257–2753/14/0324–0438$39.50/0 E-Mail [email protected] www.karger.com/ddi
ic dysplasia, though trials lack statistical significance. Liver transplantation is the only option in young patients with PSC and advanced liver disease. © 2014 S. Karger AG, Basel
Introduction
Primary sclerosing cholangitis (PSC) is a progressive cholestatic liver disease characterised by inflammation and fibrosis of both small and large bile ducts of the liver. Although sclerosing cholangitis was first described in the 1860s by Hoffman [1], only a few cases were described over the next 100 years, usually at post-mortem. However, the advent of endoscopic retrograde cholangiopancreatography (ERCP) in the 1970s enabled the diagnosis of PSC to be made ante mortem. Diagnosis was further improved by magnetic resonance cholangiopancreatography (MRCP) in the 1990s. The clinical picture can range from an incidental finding of a raised alkaline phosphatase (ALP) in an asymptomatic patient, to fatigue, lethargy and pruritis. It may progress to cirrhosis of the liver and subsequent liver failure leading to death or liver transplantation. Although PSC is considered to be a rare disease, its incidence is increasing around the world. The exact aetiopathogenesis remains unknown, but it is immune mediated, occurring in genetically predisposed individuals. There is a strong Roger W. Chapman Translational Gastroenterology Unit, Level 5 John Radcliffe Hospital, Headley Way, Headington Oxford OX3 9DU (UK) E-Mail Roger.chapman @ ndm.ox.ac.uk
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
Key Words Cholangiocarcinoma · Cholestasis · Inflammatory bowel disease · Primary sclerosing cholangitis
Epidemiology
PSC is a rare disease, which is predominantly manifested in men, with a male:female ratio of around 2: 1. Whilst the disease can present at any age, the mean age of presentation is 40 years. Prevalence rates are of the order of 0.22 per 100,000 to as high as 16.2 per 100,000 [2–4]. A recent systematic review found the incidence to be as high as 1.3 per 100,000 per year in some populations [5]. The incidence and prevalence of PSC varies according to geographic distribution, with an increasing prevalence in Northern Europe. The incidence appears to be increasing worldwide, though it is not clear whether this is a true increase, or reflects improved detection and diagnosis [6]. There is an inherited genetic predisposition to the disease, which is discussed further below, with first-degree relatives of affected patients having a 10- to 20-fold increased risk of developing the disease above that of the general population. Environmental factors are also involved with the absence of cigarette smoking appearing to be a predisposing risk factor [7, 8], which is independent of the presence of IBD. In one study, only 4.9% of patients with PSC smoked compared with 26.1% of controls [9]. There is a strong association with IBD which is dependent on geographic location. About two thirds of Northern European patients with PSC will have IBD, compared with about half of Southern Europeans, and only 20% of Japanese patients.
Aetiology
The exact aetiopathogenesis remains unknown, but it is established as being immune mediated, occurring in genetically predisposed individuals and strongly associated with IBD. The reason(s) for the close association of PSC with IBD is unclear. Theories include exposure of the liver and biliary system to antigens derived from bacteria present in the ‘biome’ through increased bowel wall permeability – the so-called ‘leaky gut’ theory. Inappropriate homing of gut-derived memory T cells to the liver has Primary Sclerosing Cholangitis
also been suggested as playing a role. These hypotheses are not mutually exclusive. Genetic Factors Many studies have investigated the genes associated with PSC. Initially, candidate gene studies were employed, followed by genome-wide association studies from Northern Europe [10]. Most recently, an international multicentre immunochip-based study was published, comparing 130,422 single nuclear polymorphisms in 3,789 patients with PSC with 25,079 controls [11]. This large study confirmed previous studies that PSC is immune related, and strongly associated to specific human leucocyte antigens (HLA) encoded on chromosome 6. Many HLA associations and other immune response genes were confirmed in the immunochip project to be common to and shared with other known autoimmune diseases, such as coeliac disease, IBD, type 1 diabetes mellitus and multiple sclerosis. Many studies have established that HLA-A1 B8 DR3, HLA-D2 and HLA-DR6 are the three genes most strongly associated with PSC, and these are evenly distributed amongst patients with PSC, regardless of whether they have IBD. It has been suggested that DR3, DR6 and DR2 encode for amino acids in the HLA β-chain that may enhance antigen presentation by the HLA molecule to the T-cell receptor. The FUT2 gene has recently been associated with PSC and Crohn’s disease, and is known to affect the behaviour of the intestinal flora [10]. Other Proposed Mechanisms Other factors are felt to play a role in aetiology, such as a toxic effect of bile on damaged biliary epithelium; the so-called ‘leaky gut’ theory, where pre-existing IBD predisposes to increased bowel wall permeability, and therefore increased exposure of the bile ducts to bacteria and other pathogens; and also an increase in homing T cells and other cytokines from the bowel to the liver, involving upregulaton of VAP-1, CCR9 and various other chemokines [12]. Descriptions of these theories are beyond the scope of this paper.
Clinical Features
Symptoms and Signs The clinical presentation of PSC is variable. Nowadays, most patients are asymptomatic, with investigation prompted by an incidental finding of a raised ALP on a routine blood test, typically in a young to middle-aged Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
439
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
association with inflammatory bowel disease (IBD), which often runs an independent course from the liver disease. It is overrepresented in men compared with women (male:female ratio of 2:1). PSC is also associated with an increased prevalence of hepatobiliary and colorectal malignancy, which accounts for a significant proportion of morbidity and mortality.
Color version available online 50 μm
Fig. 1. MRCP image showing typical features in PSC with strictur-
man with co-existing IBD. Conversely, patients may suffer from lethargy, intermittent jaundice, weight loss, right upper quadrant discomfort and pruritis. Rarely, patients present with decompensated liver disease. During the course of the disease, patients may occasionally develop symptoms of acute cholangitis, such as fever, rigors and abdominal pain, due to infection proximal to a dominant stricture. Physical examination findings are few. Hepatomegaly and splenomegaly are the most common findings in patients with advanced disease. Serology Serum biochemistry typically reveals a cholestatic picture, with ALP usually elevated to at least three times the upper limit of normal and raised γ-glutamyltranspeptidase. A normal ALP does not preclude the diagnosis if suspected. Serum aminotransferases are often mildly raised to 2–3 times the upper limit of normal, though they can be normal. Bilirubin is uncommonly raised at presentation, but often rises in the later course of the illness, representing more advanced disease. Up to 60% of patients have a mildly raised serum IgG level, up to 1.5 times the upper limit of normal, and are almost always raised in children. Around half of patients have raised IgM levels. Serum IgG4 levels are raised in 9–26% of patients with PSC, compared with only 1% of patients with another biliary disease, primary biliary cirrhosis [13–16]. It is important to test serum IgG4 levels in all patients with PSC, as elevated levels (>1.4 g/l) may confer a poorer prognosis [13, 15]. 440
Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
Fig. 2. Part of a portal tract in PSC, showing a bile duct (long arrow) with mild reactive changes in the epithelium and with concentric (‘onion skin’) fibrosis. An adjacent hepatic artery branch (short arrow) is also seen. Only a very light chronic inflammatory cell infiltrate comprising lymphocytes and plasma cells is present. HE. Magnification ×200. Photo courtesy of Adrian Bateman, Southampton, UK.
Antibodies against cystoplasmic antigens in neutrophils (ANCA) are detected in 80% of PSC patients, though it is not specific – ANCA are present in 30–40% of patients with ulcerative colitis (UC) and may potentially serve as a risk factor for developing PSC. The type of ANCA is different from those in vasculitides, which are against the proteinase PR3 and myeloperoxidase. It is unclear whether ANCA have a pathogenic role in PSC, but they may be useful to aid diagnosis.
Diagnosis
The diagnosis of PSC may be made in the setting of chronic cholestasis on biochemistry (elevated ALP and γ-glutamyltranspeptidase), with supportive cholangiographic findings. Usually, MRCP is employed as the diagnostic imaging modality of choice, with ERCP, given its associated risks, reserved only for therapeutic options and/or poor-quality MRCP (fig. 1). If the cholangiogram is normal, but PSC is suspected, a liver biopsy is indicated. Whilst the pathological findings in PSC can be highly variable, characteristic features include portal tract inflammation with lymphocytes, progressing to obliterative concentric fibrosis (so-called ‘onWilliamson /Chapman
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
ing, beading and dilatation of both intra- and extrahepatic ducts. Photo courtesy of Helen K. Bungay, Oxford, UK.
ion skinning’) and bile duct destruction (‘ductopenia’; fig. 2). The basement membrane of the bile duct is often thickened, and copper deposition is usually present. In the setting of cholestasis, a normal cholangiogram and a liver biopsy supportive of PSC, a diagnosis of smallduct PSC may be made. This confers a good prognosis for the patient, with very little chance of progressing to advanced fibrosis and liver decompensation and/or death unless it progresses to classical PSC. There have been no cases of hepatobiliary or colorectal malignancy reported in smallduct PSC. Approximately 20–25% of patients with smallduct PSC progress to classical PSC. A change in the clinical picture, such as the development of pruritis or increases in bilirubin or ALP, for example, should prompt the clinician to re-image the patient with small-duct PSC to look for cholangiographic transition to classical PSC, in which case the prognosis conferred is then that of classical PSC. The other indication for a liver biopsy in PSC is in the setting of suspicion of overlap with autoimmune hepatitis, when there are markedly elevated transaminases and/ or IgG levels, which may occur in up to 10% of cases. Secondary causes of sclerosing cholangitis must be excluded, such as previous biliary surgery, bile duct stones, toxic damage to bile ducts, e.g. intrahepatic injection of formalin, 5-fluorodeoxyuridine and alcohol, AIDS and IgG4-associated cholangitis (for which HISORt criteria may be applied) [17].
without PSC, and IBD often pre-dates the diagnosis of PSC, though not always. In one large study of a well-defined Dutch cohort of 380 patients with PSC and IBD, 75% had UC, 21% had Crohn’s disease, and 4% had IBD-unspecified [23]. Of those with UC, 83% had a pancolitis, 13% a left-sided colitis, and 4% had proctitis only. Of those with Crohn’s disease, 95% had ileocolitis, and only 5% had ileitis alone. In a subgroup analysis of 80 PSC-IBD patients compared with 80 IBD controls, 85% of PSC-IBD patients had inflammation of the right hemicolon, compared with 54% of those with IBD alone. Interestingly, in contrast to PSC-UC, the clinical activity of Crohn’s colitis in patients with PSC is similar to patients who have Crohn’s colitis without PSC [24]. The clinical activity of IBD is not related to the severity of PSC and appears to run an independent course. Having said that, there is evidence suggesting that the more severe the course of PSC, the milder the form of UC, with fewer flares and a reduced need for surgery [25]. Interestingly, when a patient undergoes liver transplantation for PSC, their IBD may worsen in severity, raising the issue of whether the presence of PSC has a protective effect on the course of IBD [26, 27]. However, a recent study contradicts this, displaying a lower index of clinical activity in those with PSC-IBD who have undergone liver transplantation compared with those who have not [28].
Osteoporosis IBD in PSC
Primary Sclerosing Cholangitis
As with other chronic liver diseases, there is an increased prevalence of osteoporosis in PSC, of the order of 15–30% [29]. Its presence does not appear to depend upon the severity of the underlying PSC [30]. European and British guidelines recommend a bone mineral density assessment by DEXA on diagnosis of PSC, as well as consideration towards supplementation with calcium and vitamin D, though this is not evidence based [31, 32].
Natural History of the Disease
Progression of Disease to Liver Failure The natural history of PSC is highly variable among patients. Observational studies have suggested the median time from diagnosis to death or liver transplantation is 10– 12 years, though in asymptomatic patients, 75% of patients survive 15 years or more. A recent population study from the Netherlands gave a much more optimistic view of a median time of 21.3 years from diagnosis to PSC-related Dig Dis 2014;32:438–445 DOI: 10.1159/000358150
441
Downloaded by: Fudan University Library 61.129.42.30 - 5/8/2015 1:24:14 AM
As discussed previously, there is a strong association of PSC with IBD. All patients diagnosed with PSC should be evaluated for IBD with a full colonoscopy if a pre-existing diagnosis of IBD does not yet exist. Around 75– 80% of the concomitant IBD is UC, with 10–15% being Crohn’s disease (predominantly colitis) and 5–10% being IBD-unspecified (colitis). Conversely, around 3–10% of patients with IBD will develop PSC, and persisting elevation in ALP in the patient with IBD (particularly pancolitis) should prompt investigation [18–20]. The phenotype of IBD in the setting of PSC has been suggested to be distinct from IBD without PSC [21, 22]. IBD is predominantly a pancolitis, and it is usually worse in severity on the right compared with the left. Backwash ileitis is more common in UC associated with PSC compared with UC alone, and the IBD is usually relatively less symptomatic than its endoscopic appearance would suggest, often running a fairly indolent course clinically. IBD is diagnosed at an earlier age in PSC, compared with IBD
100
ting of acute cholangitis related to a dominant stricture, and prophylactic antibiotics, such as low-dose ciprofloxacin, are sometimes required in the setting of recurrent cholangitis.
Population-based cohort LT centres cohort
60
Malignancy 40 20 0
Patients at risk
p < 0.001 0 590 422
5 10 15 20 25 30 Time since diagnosis until LT or PSC-related death (years) 378 266
206 143
104 67
50 26
18 9
5 0
Fig. 3. Survival until liver transplantation or PSC-related death of PSC patients in a large population cohort compared with a liver transplantation (LT) centre cohort (figure courtesy of Boonstra et al. [33]).
death or liver transplantation [33]. This may be a more realistic figure, as it represents a large population, as opposed to tertiary centres with potential referral bias (fig. 3). A well-characterised Swedish cohort of 604 patients with PSC [34] showed a mortality rate of 28% during a median follow-up of 5.7 years, with 44% of deaths due to cancer and 37% due to liver disease. 15% of the whole 604-patient cohort had undergone liver transplantation, and 23% of patients experienced cancer at some time (table 1). Levels of ALP fluctuate over the course of the illness, but about a third of patients may have a persistent reduction in ALP below 1.5 times the upper limit of normal. If this occurs, it is felt to be a positive prognostic factor, with only 3% going on to liver decompensation, liver transplantation or liver-related death [35, 36]. Dominant Strictures Patients may develop progressive jaundice, worsening of their liver biochemistry or symptoms of cholangitis, prompting investigation for a dominant stricture. A dominant stricture is defined as a stricture
