Cystic fibrosis (CF) is a multi-organ genetic disease affecting approximately 1 in every 2500 newborns worldwide. When first described in 1938, the disease was almost invariably fatal during early childhood and for many years, the basic defect has remained unknown. To date, median survival is over 30 years, and CF can no longer be considered an exclusively pediatric condition.
Understanding of the molecular and cellular basis of the disease has greatly increased following the linkage-based technique discovery of the gene responsible for CF in 1989, and the subsequent recognition of the key role of the encoded protein, the Cystic Fibrosis Transmembrane Regulator (CFTR), in maintaining fluid balance across epithelial cells (1). Absence or dysfunction of this cAMP-dependent chloride channel involved in transmembrane efflux of chloride ions, causes inability to maintain the luminal hydration of ducts, leading to thick and viscous secretions and obstruction of exocrine glands (2); this is the mechanism that mainly contributes to CF morbidity and mortality that are mainly related to lung disease.
With improved life expectancy, the clinical attention to other manifestations of the disease has increased, particularly for the wide spectrum of hepatic problems that include specific alterations ascribable to the underlying CFTR defect and lesions of iatrogenic origin or reflecting the effects of a disease process that occurs outside the liver (3) (Table 1). In this review, we shall focus on the most recent achievements regarding the pathognomonic hepatic manifestation of CF, focal biliary cirrhosis that is present in around one third of CF patients. This lesion results from biliary obstruction and progressive periportal fibrosis and is the most clinically relevant CF-associated hepatic problem, because extension of the initially focal fibrogenic process may lead to multilobular biliary cirrhosis and portal hypertension (3).
Pathogenesis of Liver Disease in CF
Current evidence suggests that CF-associated liver disease (LD) is related to the CF basic defect: expression of CFTR at the hepatobiliary level has been shown to occur exclusively at the apical domain of cholangiocytes and gallbladder epithelial cells (4). CFTR is not expressed on hepatocytes and other liver cells, and its main role is to regulate ductular secretion.
Under physiological conditions, the cAMP-stimulated Cl- secretion through low conductance Cl- channels imposes a negative luminal potential and an osmotic gradient that triggers passive secretion of Na+ and water. The change in apical Cl- gradient facilitates HCO3- extrusion via Cl-/HCO3- exchange, providing the biliary alkalinization required for the digestive function and for the solubility of organic components of bile. In CF patients, absence or dysfunction of CFTR may reduce bile fluidity and alkalinity, leading to plugging of intrahepatic bile ducts by inspissated biliary secretions. An association between cirrhosis and bile duct obstruction by mucus plugs was reported in an autopsy study (5), however, the morphological aspects of bile duct obstruction may not be evident after the neonatal period.
Injury of bile duct cells is probably the primary event in the development of periportal fibrosis, as suggested by the ultrastructural abnormalities of cholangiocytes, consistently documented in CF patients, with irregular shapes, necrosis and periductular collagen deposition (6); these lesions may represent the anatomical expression of long-standing impairment in ductular bile flow, that in turn may increase susceptibility of the biliary epithelium to damage by detergents in the form of endogenous bile acids and to injury by infectious agents. Indeed, CF-associated LD is currently considered the first inherited LD resulting from impaired secretory function of the biliary epithelium. However, it remains to be explained why only one third of CF patients develop LD and why LD shows great degree of variability in terms of severity. Only in a minority of patients, often in the pediatric age, LD may represent the main clinical problem and its progression may be unusually rapid.
Although development of LD seems to be restricted to patients with a severe genotype, no specific CFTR mutations has been associated with the presence and severity of LD (7,8), suggesting a multifactorial pathogenesis (Table 2). The role of environmental factors is suggested by the occasional familial clustering of LD; factors related to poor nutritional status, antioxidant deficiency, noncompliance with treatment and chronic drug hepatotoxicity may worsen the biliary secretory defect and liver damage. This may also occur as a result of concomitant events such as viral hepatitis, abdominal surgery and total parenteral nutrition. On the other hand, discordant liver expression in CF siblings in whom environmental factors are presumably similar has suggested that the clinical expression and severity of LD in CF may be modulated by genetic factors inherited independently from the CFTR gene (9). Preliminary data regarding genetic modifiers for CF-associated LD suggest a complex multigenic inheritance with possible interaction of different candidate genes (eg. α1 anti-trypsin deficiency, TGF-beta cytokine, mannose-binding lectin 2 and glutathione S-transferase) (Table 2); identification of genetic modifiers is a relevant issue because it may allow identification of patients at risk for the development of LD at the time of diagnosis of CF and early institution of prophylactic strategies.
Frequency, Natural History and Clinical Features
Autopsy studies have documented a progressive increase in prevalence of LD with age, from 10% in infants to 72% in adults (10); prevalence figures obtained by retrospective analyses of clinical records of CF patients are much lower, ranging between 4.2% (11) and 17% (12). Although in the past, attention was focused exclusively on the small proportion (∼5%) of patients in whom LD had progressed to end-stage multilobular biliary cirrhosis who were the only ones who received therapeutic interventions, the last decade has been characterized by the active search for LD and increased recognition of asymptomatic patients with focal biliary cirrhosis. Indeed, cross-sectional studies in which LD was actively searched using biochemical and ultrasonographic assessment have provided higher prevalence figures (from 18 to 37%) (10), and prospective studies have reported cumulative incidence of LD ranging from 27 to 35% without incident cases after the age of 18 (13,14). Thus, LD is a relatively early complication of CF-in the majority of affected patients it becomes clinically apparent by the end of the first decade of life, but may develop earlier, particularly in CF patients with a positive history of meconium ileus (14). Although rarely, CF may present with a picture of neonatal cholestatic jaundice mimicking biliary atresia because of obstruction of extrahepatic bile ducts by viscous biliary secretions. This is also a direct consequence of the basic CF defect, but generally resolves spontaneously within the first months of life and does not progress to clinically relevant LD (15).
CF-associated LD generally displays a slowly progressive course, with hepatocellular failure being a late event, whereas development of portal hypertension and related complications occurs earlier and more frequently (12,16). In the prospective study carried out at our Center (14), hepatic status was strictly monitored in 177 CF patients for a median follow-up period of 14 years (2432 patient-years). Forty-eight patients developed LD, with liver cirrhosis already present in 5; the incidence of cirrhosis was 4.5% (95% CI: 2.3-7.8) during a median period of 5 years from diagnosis of LD and incidence of major complications of cirrhosis (esophageal varices, gastrointestinal bleeding, ascites) was 0.4 cases per 100 cirrhotic patient-years (95% CI: 0-2). Hepatic synthetic function was sufficiently maintained in the majority of cirrhotic patients, and only 1 liver transplantation was required. Development of LD did not predict different mortality (death rate ratio: 0.4; 95% CI: 0.1-1.5) or a higher incidence of other clinically relevant outcomes. Considering that a liver decompensation rate of 10% generally characterizes liver cirrhosis in adults (17), these data and those obtained by other prospective studies suggest that the evolution of CF-associated LD is more favorable than estimated in the past (12) and that until the advanced stages are attained, its presence does not influence the severity of other clinical manifestations of CF (13,14,18).
Impact of End-stage Liver Disease on CF
The impact of advanced LD on the pulmonary function and the nutritional status of CF patients is becoming increasingly evident. During the progression of LD, CF patients are at risk of developing several extrahepatic complications, including malnutrition, hepatic osteodystrophy and deterioration of pulmonary status. A recent case-control study reported that CF-patients with advanced LD were shorter, lighter and had reduced fat mass compared to those without LD (19). The pathogenesis of malnutrition in CF patients with end-stage LD is multifactorial and involves the increase in resting energy expenditure characteristic of the disease, abnormalities in nutrient intake (because of anorexia and in some patients to protein restriction), fat malabsorption (because of the combined effect of cholestasis and pancreatic insufficiency) and abnormal metabolism of nutrients. In this respect, it should be noted that CF patients more often develop end-stage LD during adolescence, when glucose intolerance or diabetes mellitus also frequently develop. An association between LD and diabetes has been recently documented within the Epidemiologic Study of CF (20) and may be related to peripheral insulin resistance.
CF patients with LD are also at risk of developing osteoporosis. In a study aimed at establishing the effects of end-stage LD on body composition, we documented severe osteopenia in a group of patients with multilobular cirrhosis and portal hypertension that was significantly corrected after transplantation, but not after conservative management (21).
Overall, the available data seem to indicate that nutritional status is not a causal factor of CF associated LD (18), but rather that CF patients with LD are at increased risk for deterioration of nutritional status. Adequate attention to nutritional needs may therefore be very important to slow the progression of LD. Cirrhosis and portal hypertension can also negatively affect respiratory function because of organomegaly, ascites-induced diaphragmatic splinting and intrapulmonary shunting, leading to recurrent respiratory infections from multiresistant bacteria, frequent hospital admissions and deterioration of quality of life (21).
Risk Factors for Liver Disease in CF
Factors found to be significantly associated with the development of CF-associated LD include male sex, history of meconium ileus, pancreatic insufficiency and severe genotype (7,8,14,22), however, their role could not be consistently confirmed. The preponderance of male subjects among CF patients with LD (11,12,16) suggests that endocrine factors may exert a protective effect on female CF patients, who in our prospectively followed cohort did not develop LD after the age of eleven years (14). We also found that incidence of LD was significantly higher in patients with a positive history of meconium ileus, most of whom had developed LD within early during the first decade of life. This association, linking inspissated gut content and biliary secretions, was first described by a necropsy study (5), but was not consistently found in cross-sectional studies involving different CF patient populations (8,13) and a study reported that meconium ileus was even associated with absence of LD (19). In patients with a positive history of meconium ileus, additional risk factors for development of LD may include abdominal surgery with extensive small bowel resection, poor nutrition in early life and prolonged total parenteral nutrition (15,23). A number of still unknown factors, including genetic modifiers associated with gender or ethnic group, may confer patients different susceptibility to liver involvement. The existence of a modifier locus for meconium ileus (CFM1) on human chromosome 19q13.2 has been reported (24) and a potential candidate is the KCNN4 gene encoding for an intermediate conductance Ca2+ activated potassium channel.
Diagnosis and Clinical Follow-Up
Evidence of LD in CF patients is often subclinical until pathological changes are diffuse and severe, therefore it is often underdiagnosed. The most common presentation is the occasional finding of hepatomegaly on routine physical examination, often but not always associated with abnormalities of liver biochemistry. Jaundice is limited to patients with neonatal cholestasis and with end-stage multilobular biliary cirrhosis. Early detection can be difficult, but should be pursued because only early lesions are likely to be reversible. Because a sensitive and specific test to evaluate biliary cell function is not yet available (25), diagnosis of CF-associated LD is based on regular clinical examination (with measurement of liver span at the mid-clavicular line) and on a combination of biochemical tests and imaging techniques. Table 3 summarizes the characteristics and limits of investigations used for diagnosis and follow-up of CF-associated LD. Biochemical abnormalities are frequently mild or intermittent and have shown low sensitivity and no correlation with histological findings (26); not infrequently CF patients with multilobular biliary cirrhosis have normal liver biochemistry. On the other hand, isolated elevation of serum transaminase levels, with normal concentrations of enzymes related to cholestasis, suggests the presence of steatosis, the most common hepatic lesion in CF, that may have a iatrogenic origin and should be adequately recognized and followed-up, after correction of nutritional deficiencies, if present. It should also be noted that nonspecific biochemical abnormalities have been documented in more than 50% of infants with CF, with complete normalization in the majority within 2-3 years of age and no impact on future development of LD (13).
Histological assessment has been so far considered of questionable accuracy in CF because of patchy distribution of lesions, and liver biopsy only rarely has been included in the diagnostic work-up of CF patients (26). Nevertheless, liver biopsy may provide important information on the type of the predominant lesion (steatosis or focal biliary cirrhosis), the extent of portal fibrosis, rate of progression of LD and response to treatment. It should be performed if there is doubt about the diagnosis, and also to establish cirrhosis or prior to treatment or liver transplantation.
With regard to imaging techniques, ultrasonography with Doppler measurements of the hepatic flow, is accurate in distinguishing normal parenchyma, from fatty infiltration, fibrosis, cirrhosis, portal hypertension or ductal abnormalities. Abnormal echogenicity frequently precedes clinical and biochemical manifestations of LD, suggesting that routine ultrasonography may be a valuable marker of early LD in CF (18,27).
Hepatobiliary scintigraphy with iminodiacetic acid derivatives can document a typical picture of biliary drainage impairment, with dilation of intra- and extrahepatic bile ducts and delayed biliary excretion and intestinal appearance of the tracer. Scintigraphy can document time-related progression of LD (28) and has been employed to monitor the response to treatment with ursodeoxycholic acid (29).
Endoscopic retrograde cholangiography (ERCP), an invasive procedure with considerable potential for complications, has been used in patients with symptoms and signs suggestive of sclerosing cholangitis, distal stenosis of the common bile duct (30) and for those with choledocholithiasis, for which it also provides a therapeutic option.
The high rate of abnormalities reported at magnetic resonance cholangiography imaging of the biliary tree in CF patients suggests that such technique may be useful for early detection of intrahepatic biliary tract involvement (31).
Hepatic Follow-Up in CF Patients
Regular monitoring of hepatic status should be accomplished in all patients with CF, with liver biochemistry and ultrasound scan included in their routine annual schedule. A simple echographic scoring system based on coarseness of liver parenchyma, nodularity of the liver edge and increased periportal echogenicity is available for the hepatic follow-up of these patients (32,33). In patients with cirrhosis, it seems prudent to determine alpha-feto protein levels annually to monitor the possible development of hepatocellular carcinoma (34). The development of esophageal varices should be evaluated by means of upper digestive endoscopy in patients with splenomegaly and hypersplenism.
Management of Patients With Liver Disease
Because of the decreasing mortality from extrahepatic causes, management of LD in CF patients is becoming a relevant clinical issue. There is potential for several therapeutic strategies, including liver targeted gene therapy and pharmacological correction of defective ion transport function. However, these approaches are still in the experimental phase. Extracellular adenosine 5-triphosphate (ATP) is increasingly recognized as a potent Cl- secretagogue and can activate the Ca2+-dependent Cl- channel in different cell types, including cholangiocytes (35), Exogenous administration of nucleotide triphosphate may therefore correct the defective anion secretion in intrahepatic ducts and represent a potential pharmacological approach to CF-associated LD. The ability of glibenclamide, a sulfonylurea with known CFTR inhibitor activity, to paradoxically stimulate cholangiocyte secretion through exocytosis in CFTR defective mice, has recently suggested that drugs with similar mechanism of action may be designed for the treatment of CF-associated and other cholestatic LDs (36).
Bile Acid Therapy
Oral bile acid therapy, aimed at improving biliary secretion in terms of bile viscosity and bile acid composition, is currently the only available therapeutic approach for CF-associated LD and is free of serious side effects (Table 4). Ursodeoxycholic acid (UDCA) administration may exert beneficial effects through the replacement of hydrophobic endogenous bile retained in cholestasis, direct cytoprotection on biological membranes, or stimulation of Cl- secretion through Ca2+-dependent Cl- channel (37,38). The optimal daily dose of UDCA (20 mg/kg body weight) is higher than that conventionally used in other cholestatic LDs (39), probably because of poor intestinal bile acid absorption in CF patients. Treatment with UDCA is widely employed in these patients after reports of beneficial effects on liver biochemistry, hepatic excretory function and biliary drainage, liver histology and nutritional and essential fatty acid status. However, its impact on the natural history of LD remains to be defined (40). In view of the fact that UDCA should be administered on a regular and probably life-long basis in CF patients, its long-term effectiveness on clinically relevant outcomes should be further investigated. Asymptomatic patients with early-stage LD are more likely to benefit from UDCA administration. Identification of CF patients at risk for the development of LD is needed to evaluate if UDCA may have a role also for the prevention of LD.
Treatment of Portal Hypertension and Hypersplenism
For more advanced LD, severe portal hypertension and variceal bleeding may require sclerotherapy and vasopressin administration during the acute episode; variceal ligation may be a preferable alternative, but both procedures must be repeated and do not solve the problem of hypersplenism. Elective surgical porto-systemic shunt was performed in patients with preserved liver function and allowed prolonged postoperative survival (16). Other therapeutic interventions include transjugular portosystemic shunt (TIPS) for portal decompression in patients waiting for liver transplantation and partial splenectomy with conservation of the upper pole of the spleen for patients with massive splenic enlargement (41). The efficacy of β-blockers has not been evaluated in CF because of their potential adverse effects on airway reactivity.
Liver transplantation is indicated for CF patients with liver failure and/or life-threatening sequelae of portal hypertension, who also have mild pulmonary involvement that is expected to support long-term survival. The 1-year survival rate is approximately 80%, with beneficial effects on lung function, nutritional status, body composition and quality of life in most cases (21). The perspective of liver transplantation should be considered early, before a critical deterioration of lung function is reached. Specific clinical scores that consider not only features of portal hypertension, hepatocellular function and hypersplenism, but also nutritional and pulmonary status are useful to evaluate the need and timing for liver transplantation in CF patients (42,43).
LD is a relatively frequent and early complication of CF. The pathogenesis seems to be multifactorial with variable contribution of environmental and genetic determinants. With improved life expectancy of CF patients in future years, the impact of LD on quality of life and survival is going to increase. UDCA is to date the only accepted treatment for LD in CF.
1. Rowe SM, Miller S, Sorscher EJ. Mechanisms of disease. Cystic Fibrosis
. N Eng J Med
2. Mehta A. CFTR: More than just a chloride channel. Pediatr Pulmonol
3. Colombo C, Battezzati PM. Liver involvement in Cystic Fibrosis
: primary organ damage or innocent bystander? J Hepatol
4. Kinnman N, Lindblad A, Housset C, et al. Expression of cystic fibrosis
transmembrane conductance regulator in liver tissue from patients with cystic fibrosis
5. Maurage C, Lenaerts C, Weber AM, et al. Meconium ileus and its equivalent as a risk factor for the development of cirrhosis: an autopsy study in Cystic Fibrosis
. J Pediatr Gastroenterol Nutr
6. Lindblad A, Hultcrantz R, Strandvik B. Bile duct destruction and collagen deposition: a prominent ultrastructural feature of the liver in Cystic Fibrosis
7. Colombo C, Apostolo MG, Ferrari M, et al. Analysis of risk factors for the development of liver disease
associated with cystic fibrosis
. J Pediatr
8. Wilschanski M, Rivlin J, Cohen S, et al. Clinical and genetic risk factors for CF-related liver disease
9. Castaldo G, Fuccio A, Salvatore D, et al. Liver expression in cystic fibrosis
could be modulated by genetic factors different from the cystic cibrosis transmembrane regulator genotype. Am J Med Gen
10. Colombo C, Battezzati PM, Podda M. Hepatobiliary disease in Cystic Fibrosis
. Sem Liv Dis
11. Scott-Jupp R, Lama M, Tanner MS. Prevalence of liver disease
in Cystic Fibrosis
. Arch. Dis Child
12. Feigelson J, Anagnostopoulos C, Poquet M, et al. Liver cirrhosis in Cystic Fibrosis
. Therapeutic implications and long-term follow-up. Arch Dis Child
13. Lindblad A, Glaumann H, Strandvik B. Natural history of liver disease
in Cystic Fibrosis
14. Colombo C, Battezzati PM, Crosignani A, et al. Liver disease
in Cystic Fibrosis
: a prospective study on incidence, risk factors and outcome. Hepatology
15. Shapira R, Hadzic R, Francavilla R, et al. Retrospective review of cystic fibrosis
presenting as infantile liver disease
. Arch Dis Child
16. Debray D, Lykavieris P, Gauthier F, et al. Outcome of cystic fibrosis
-associated liver cirrhosis: management of portal hypertension. J Hepatol
17. Sherlock S, Dooley J. Hepatic cirrhosis. Diseases of the Liver and Biliary System
, 10th. Oxford: Blackwell Science, 1997:371-84.
18. Ling SC, Wilkinson JD, Hollman AS, et al. The evolution of liver disease
in Cystic Fibrosis
. Arch Dis Child
19. Corbett K, Kelleher S, Rowland M, et al. Cystic fibrosis
-associated liver disease
: a population-based study. J Pediatr
20. Marshall BC, Butler SM, Stoddard M, et al. Epidemiology of cystic fibrosis
-related diabetes. J Pediatr
21. Colombo C, Costantini D, Rocchi A, et al. Effects of liver transplantation on the nutritional status of patients with Cystic Fibrosis
. Transplant International
22. Lamireau T, Monnereau S, Martin S, et al. Epidemiology of liver disease
in cystic fibrosis
: a longitudinal study. J Hepatol
23. Lai Hc, Kosorok MR, Losova A, et al. Nutritional status of patients with cystic fibrosis
with meconium ileus: a comparison with patients without meconium ileus and diagnosed early through neonatal screening. Pediatrics
24. Zielenski J, Corey M, Rozmahel R, et al. Detection of a cystic fibrosis
modifier locus for meconium ileus on human chromosome 19q13. Nat Genet
25. Narkewicz MR. Markers of Cystic Fibrosis
- associated liver disease
. J Pediatr Gastroenterol Nutr
26. Potter CJ, Fishbein M, Hammond S, et al. Can the histologic changes of Cystic Fibrosis
-associated hepatobiliary disease be predicted by clinical criteria? J Pediatr Gastroenterol Nutr
27. Lenaerts C, Lapierre C, Patriquin H, et al. Surveillance for cystic fibrosis
-associated hepatobiliary disease: early ultrasound changes and predisposing factors. J Pediatr
28. Foster JA, Ramsden WH, Conway SP, et al. The role of IDA scintigraphy in the follow-up of liver disease
in patients with cystic fibrosis
. Nucl Med Commun
29. Colombo C, Castellani MR, Balistreri WF, et al. Scintigraphic documentation of an improvement in hepatobiliary excretory function after treatment with ursodeoxycholic acid in patients with Cystic Fibrosis
and associated liver disease
30. Gaskin KJ, Waters DLM, Howman-Giles R, et al. Liver disease
and common bile-duct stenosis in Cystic Fibrosis
. N Eng J Med
31. Durieu I, Pellet O, Simonot L, et al. Sclerosing cholangitis in adults with cystic fibrosis
: a magnetic resonance cholangiographic prospective study. J Hepatol
32. Williams SGJ, Evanson JE, Barrett N, et al. An ultrasound scoring system for the diagnosis of liver disease
in Cystic Fibrosis
. J Hepatol
33. Fagundes E, Silva R, Roquete M, et al. Validation of the Williams ultrasound scoring system of liver disease
in cystic fibrosis
. J Pediatr (Rio J)
34. Mckeon D, Day A, Parmar J, et al. Hepatocellular carcinoma in association with cirrhosis in a patient with cystic fibrosis
. J Cyst Fibros
35. Roman RM, Feranchak A, Salter KD, et al. Endogenous ATP regulates Cl-
secretion in cultured human and rat biliary epithelial cells. Am J Physiol
36. Spirli C, Fiorotto R, Song L, et al. Glibenclamide stimulates fluid secretion in rodent cholangiocytes through a CFTR-independent mechanism. Gastroenterology
37. Paumgartner G, Beuers U. Ursodeoxycholic acid in cholestatic liver disease
: Mechanisms of action and therapeutic use revisited. Hepatology
38. Chinet T, Fouassier L, Dray-Charier N, et al. Regulation of electrogenic anion secretion in normal and Cystic Fibrosis
gallbladder mucosa. Hepatology
39. Colombo C, Crosignani A, Assaisso ML, et al. Ursodeoxycholic acid therapy in Cystic Fibrosis
associated liver disease
: a dose-response study. Hepatology
40. Cheng K, Ashby D, Smyth R: Ursodeoxycholic acid for Cystic Fibrosis
-related liver disease
(Cochrane Review). The Cochrane Library. Issue 2. Oxford: Update Software6-12, 2000
41. Zach MS, Thalhammer GH, Eber E. Partial splenectomy in CF patients with hypersplenism. Arch Dis Child
42. Noble-Jamieson G, Barnes N, Jamieson N, et al. Liver transplantation for hepatic cirrhosis in cystic fibrosis
. J Royal Soc Med
43. Milkiewicz P, Skiba G, Kelly D, Weller P, et al. Transplantaion for Cystic Fibrosis
: outcome following early liver transplantation. J Gastroenterol Hepatology