Bruce Y. Tung

Ursodiol Use Is Associated with Lower Prevalence of Colonic Neoplasia in Patients with Ulcerative Colitis and Primary Sclerosing Cholangitis

Patients with ulcerative colitis have an increased risk for colorectal neoplasia; this risk approaches 0.5% to 1% per year of disease. The extent and duration of colitis, as well as age at onset of disease, are important risk factors for neoplastic progression (1, 2). Primary sclerosing cholangitis is a chronic inflammatory disease of the biliary tract that occurs in 2% to 4% of patients with ulcerative colitis (3-5). Although the data are still controversial, most studies have shown that patients with ulcerative colitis and primary sclerosing cholangitis are at even higher risk for colonic neoplasia than are those with ulcerative colitis alone (6-11). The risk for colonic dysplasia or cancer in patients with ulcerative colitis and primary sclerosing cholangitis approaches 50% after 25 years of colitis (6-8). Epidemiologic studies suggest that environmental and dietary factors are important in the development of colorectal cancer (12). High-fat diets predispose to colorectal cancer, in part because they generate tumor-promoting secondary bile acids (13, 14). Conversely, the use of aspirin and other nonsteroidal anti-inflammatory drugs is associated with reduced risk for colorectal cancer and adenomatous polyps (15-17). Controversy surrounds the potential chemoprotective role of sulfasalazine in patients with ulcerative colitis (18, 19). Ursodiol, the 7--epimer of chenodeoxycholic acid, has been used in patients with primary sclerosing cholangitis because it substantially improves biochemical indices of liver function, although overall disease progression does not appear to be affected (20). Ursodiol use reduces the colonic concentration of the secondary bile acid deoxycholic acid (21). Although experimental evidence in animal models suggests that ursodiol administration inhibits colonic carcinogenesis, no clear evidence of inhibitory activity has been demonstrated in human studies (7, 22, 23). We sought to determine whether ursodiol use is associated with lower risk for colonic dysplasia, the precursor to colon cancer, in patients with ulcerative colitis and primary sclerosing cholangitis. Methods Patients We studied 59 patients with ulcerative colitis and primary sclerosing cholangitis who were enrolled in the colonoscopic surveillance program at the University of Washington, Seattle, Washington. These patients were involved in a research study of colonic surveillance in ulcerative colitis according to a protocol approved by the Human Subjects Division of the University of Washington. Forty-three patients were male and 16 were female. The diagnosis of primary sclerosing cholangitis was based on conventional cholangiographic criteria. Confirmatory histologic findings in the 35 patients who underwent liver biopsy included concentric periductal fibrosis, destructive inflammatory lesions of ducts, changes of ductal obstruction, or ductopenia ( 50% of at least 20 portal tracts without a duct). The diagnosis of ulcerative colitis was made according to standard histologic criteria: presence of basal lymphoplasmacytosis and distortion of crypt architecture, defined by the presence of two or more branched or irregularly shaped crypts in a single mucosal biopsy specimen. Clinical Variables Clinical variables were abstracted from the medical record and patient interview. Study groups were defined according to whether patients had ever used ursodiol. For all patients whose medical records failed to mention ursodiol use, confirmation was made by direct patient interview. All medications were carefully noted: specifically, the dates of use of ursodiol, sulfasalazine, other 5-aminosalicylic acid preparations, prednisone, azathioprine, cyclosporine, and methotrexate. Estrogen use and menstrual status were assessed in women. Age at onset of symptomatic colitis and date of diagnosis of sclerosing cholangitis were also noted. To estimate the severity of liver disease, ChildPugh class was calculated at the time of each surveillance colonoscopy. For patients who had previously undergone orthotopic liver transplantation, a separate classification was created to reflect the fact that they had had severe liver disease despite a normal current ChildPugh score. Surveillance Colonoscopy for Diagnosis of Dysplasia Colonoscopy was performed every 3 years unless indefinite or low-grade dysplasia or aneuploidy was detected, in which case-patients underwent yearly colonoscopy. Patients who had high-grade dysplasia were referred for colectomy. At colonoscopy, biopsy samples were obtained from flat mucosa (by using a large cupped biopsy forceps) from four quadrants at 10-cm intervals from the cecum through the descending colon and at 5-cm intervals in the sigmoid colon and rectum. Any mucosal irregularity or polyp was sampled separately. The mean number of biopsy samples obtained per colonoscopy was 46.8 1.5. The biopsy samples were oriented submucosal-side-down on monofilament plastic mesh and placed in Hollande fixative. All biopsy samples were step-serial sectioned, stained with hematoxylin and eosin, and evaluated by an experienced gastrointestinal pathologist who had no knowledge of the clinical history or colonoscopic findings. The histologic criteria for the diagnosis and grading of dysplasia established by the Inflammatory Bowel Disease/Dysplasia Morphology Study Group were applied, except that the category indefinite for dysplasia was not subdivided (24). Statistical Analysis Study groups were defined according to whether the patient had ever used ursodiol. The primary outcome studied was occurrence of colonic dysplasia. A patient was defined as having achieved the outcome if he or she was known to have dysplasia at any point up to the last surveillance. Hence, the fraction of patients with the dysplasia outcome represented the prevalence of ever having had a diagnosis of dysplasia. For the primary analysis, patients whose biopsies were read as indefinite for dysplasia were combined with those whose biopsies were negative for dysplasia. Two secondary analyses were performed. In the first, all patients classified as indefinite for dysplasia were excluded from the analysis. In the second, the outcome was changed to high-grade dysplasia rather than low-grade dysplasia and high-grade dysplasia. For all analyses, ursodiol use and other variables were assessed up to the time of the outcome or last surveillance. Other variables assessed were use of sulfasalazine, other 5-aminosalicylic acid preparations, prednisone, cyclosporine, azathioprine, or methotrexate; duration of sclerosing cholangitis; duration of ulcerative colitis; ChildPugh class; duration of ursodiol use; age at onset of colitis; and sex. Unpaired t-tests and the Fisher exact test were used to compare means and proportions, respectively. Unadjusted odds ratios were estimated directly from 2 2 tables for ursodiol use versus outcome status, and adjusted odds ratios were estimated by using logistic regression. MathSoft S-Plus 3.4 (MathSoft, Inc., Seattle, Washington) and GraphPad Prism 2.01 (GraphPad Software, San Diego, California) statistical software packages were used. Role of the Funding Source The funding source had no role in the collection, analysis, or interpretation of the data or in the decision to submit the study for publication. Results Fifty-nine patients with ulcerative colitis and primary sclerosing cholangitis were enrolled in the colonoscopic surveillance program. Forty-one patients (69%) had used ursodiol, and 18 (31%) had never used ursodiol. Characteristics of ursodiol users and nonusers are shown in Table 1. Twenty-six patients (44%) had a diagnosis of dysplasia at some time during surveillance; no patient developed cancer. Table 1. Patient Characteristics Of the 26 patients who had dysplasia, 50% (13 of 26) had used ursodiol compared with 85% (28 of 33) of patients with no dysplasia. This difference in ursodiol use was highly significant, with an odds ratio of 0.18 (CI, 0.05 to 0.61; P=0.005) (Table 2). The negative association between dysplasia and ursodiol use remained after adjustment for sex, age at onset of ulcerative colitis, duration of ulcerative colitis, duration of sclerosing cholangitis, ChildPugh class, and use of sulfasalazine (adjusted odds ratio, 0.14 [CI, 0.03 to 0.64]; P=0.01) (Table 3). Results were similar when treatment with 5-aminosalicylic acid, prednisone, cyclosporine, azathioprine, or methotrexate and number of surveillance colonoscopies were included in the model. Because a large number of covariates were included in the final logistic regression model relative to the sample size of 59, computer simulations were performed to check the stability of the modeling procedure. Both the Pvalue and the 95% confidence interval were reliable when the true odds ratio was assumed to be 0.14 to 1.0. Table 2. Medication Use and Development of Dysplasia Table 3. Multivariate Analysis of Risk Factors for Dysplasia Among patients who used ursodiol, the mean duration of use was 3.5 years for those who progressed to dysplasia compared with 4.2 years for those who did not progress to dysplasia (P>0.2). The mean dosage was 9.9 0.7 mg/kg per day in those who progressed to dysplasia and 8.9 0.5 mg/kg per day in those who did not progress to dysplasia (P>0.2). The relationship between total accumulated ursodiol dose (dose duration of use) and progression to dysplasia was not significant (P>0.2). Sixteen of 59 patients (27%) were indefinite for dysplasia at their most advanced colonic histologic stage. In a secondary analysis that excluded these 16 patients, ursodiol use was still negatively associated with colonic dysplasia (odds ratio, 0.21 [CI, 0.05 to 0.99]; P=0.05). After controlling for sex, age at onset of colitis, duration of colitis, duration of sclerosing cholangitis, ChildPugh class, and use of sulfasalazine, the adjusted odds ratio was 0.20 (CI, 0.03 to 1.5; P=0.12). Eleven of the 59 patients (19%) had a diagnosis of high-grade dysplasia duri

Biliary complications of orthotopic liver transplantation.

Biliary complications are a common cause of morbidity following orthotopic liver transplantation. Complications involving the biliary tree occur after 6–34% of all liver transplants performed, usually within the first 3 months after transplantation. Bile leaks and biliary strictures are the most common biliary complications, but sphincter of Oddi dysfunction, hemobilia, and biliary obstruction from stones, sludge, or casts have also been described. The risk of specific biliary complications is related to the type of biliary reconstruction performed at the time of transplantation. In this article, we review the major types of biliary reconstruction and their associated biliary complications. Specific risk factors for the development of biliary complications are outlined. Finally, the management of biliary complications is discussed, with an emphasis on the role of endoscopic therapy.

Cholestasis and alcoholic liver disease.

Histologic cholestasis and clinical jaundice may be seen in all stages of alcoholic liver disease. In rare cases, isolated cholestasis without significant steatosis, hepatitis, or cirrhosis is identified in an alcoholic patient. The mechanisms of ethanol-induced cholestasis are not well studied but may involve compression of intrahepatic biliary radicals or interference with basolateral uptake and intracellular transport of bile acids. In the evaluation of the jaundiced alcoholic patient, clinical, biochemical, and radiologic data are usually sufficient to distinguish alcohol-induced liver disease from extrahepatic biliary obstruction. In cases where the diagnosis is not readily apparent, more invasive studies such as liver biopsy or ERCP may be necessary. The risk of these invasive studies is directly related to the degree of underlying hepatic dysfunction.

Endotheliitis in chronic viral hepatitis: a comparison with acute cellular rejection and non-alcoholic steatohepatitis.

Endotheliitis is an important histologic feature of acute cellular rejection (ACR) in the liver allograft. This change is not specific, however, and has been suggested to be associated with various liver diseases. End-stage liver disease owing to chronic hepatitis C is the leading indication for transplantation in North America, and its recurrence in allograft recipients is common. Because the presence of endotheliitis remains a diagnostic and therapeutic dilemma in transplant pathology, we investigated the prevalence and severity of endotheliitis in chronic liver diseases including hepatitis C. Endotheliitis was evaluated in 128 nontransplant liver biopsies of chronic liver diseases before therapy, including hepatitis C (HCV, n=62), hepatitis B (HBV, n=17), and nonalcoholic steatohepatitis (NASH, n=49). Eighty posttransplant biopsies with ACR were also reviewed. Subendothelial and supraendothelial endotheliitis were separately scored in the portal and central regions using a semiquantitative scoring system from 0 to 4. Pathologists were blinded to the clinical histories, and each biopsy was independently scored by 2 pathologists. Histologic activity index was also scored subsequently for cases of chronic HCV and HBV, using the modified Knodell (Ishak) score. Mean endotheliitis scores >1 were seen in 60%, 35%, and 6% of HCV, HBV, and NASH patients, respectively. The scores for portal subendotheliitis and supraendotheliitis were significantly higher in the viral hepatitis group than in the NASH group (P<0.01). There was no significant difference in the scores of endotheliitis comparing HCV to HBV. ACR group showed significantly higher scores in both portal and central subendotheliitis than any other group (P<0.00005). In the HBV and HCV groups with mean scores of portal subendotheliitis >1 (n=44), mean Ishak scores for portal inflammation and periportal injury were 2.43 and 2.34, respectively; whereas in those with less severe portal subendotheliitis (≤1, n=35), Ishak scores were 1.66 and 1.37, respectively (P=0.00001 for portal inflammation and P=0.00001 for periportal injury, respectively). Our results suggest that minimal to mild subendotheliitis is common in portal veins in chronic hepatitis C and B, but is significantly less intense than that seen in ACR. The degree of endotheliitis correlates with inflammatory activity. These observations may help minimize the risk of overdiagnosing ACR when the patient has recurrent viral hepatitis, and may help clinicians avoid exposing patients to unnecessary immunosuppressive regimens when patients do not have cellular rejection.

Hepatitis B and Liver Transplantation

Liver transplantation is the treatment of choice for patients with liver failure secondary to chronic hepatitis B. However, liver transplantation is complicated by the risk of recurrent hepatitis B virus infection, which significantly impairs graft and patient survival. The main risk factor for the development of recurrent hepatitis B virus infection is the virus load at the time of transplantation. The development of antiviral medications, such as lamivudine and adefovir, and the implementation of effective prophylactic regimens using hepatitis B immune globulin have significantly improved the outcomes of hepatitis B after liver transplantation. However, current approaches continue to be hampered by the extremely high cost of treatment and the emergence of drug-resistant viral mutations. Ongoing studies are necessary to establish the most cost-effective approaches to prevent recurrent hepatitis B virus infection after liver transplantation.

CLINICAL MANAGEMENT OF IRON OVERLOAD

HHC is a common inherited disorder, characterized by iron accumulation in the liver, heart, pancreas, and other organs. The clinical consequences of systemic iron loading are diverse and not always improved with iron reduction therapy. The most important prognostic factor at the time of diagnosis is the presence or absence of hepatic fibrosis or cirrhosis. Those without significant hepatic fibrosis may be expected to have a normal life expectancy with phlebotomy therapy. The availability of genetic testing for HHC has significantly changed the diagnostic approach to this disorder. Although liver biopsy remains vital to determining prognosis, genetic testing is increasingly used in the diagnosis and family screening of patients with HHC.

Human Liver Cytochrome P450 2D6 Genotype, Full-length Messenger Ribonucleic Acid, and Activity Assessed with a Novel Cytochrome P450 2D6 Substrate

The goal of this study was to develop and validate a cytochrome P450 (CYP) 2D6 probe substrate with improved sensitivity to elucidate the relationship of CYP2D6 ribonucleic acid transcript levels, genotype, and enzyme activity in human liver biopsy samples.

LIVER TRANSPLANTATION FOR HEMOCHROMATOSIS, WILSON'S DISEASE, AND OTHER METABOLIC DISORDERS

Liver transplantation provides an effective means for replacing a failing liver, in addition to correcting the underlying abnormality in many metabolic disorders. Results of liver transplantation for metabolic diseases have been generally encouraging, with the exception of hereditary hemochromatosis, in which infectious and cardiac complications appear to increase post-transplant mortality. Better pretransplant diagnosis of hemochromatosis, utilizing the recently identified putative gene, may help reduce post-transplant complications. In metabolic diseases, improved understanding of the underlying genetic and molecular defects will lead to advances in medical therapy and perhaps a decreased need for liver transplantation. NTBC therapy for hereditary tyrosinemia and purified glucocerebroside therapy for Gaucher disease are two such examples. The prospects of gene therapy are being actively pursued for many metabolic diseases, such as CF, hemophilia, and familial hypercholesterolemia. Until such investigation leads directly to clinical practice, however, liver transplantation remains an effective option for therapy for a wide range of metabolic diseases.