HEPATOLOGY WATCH®
Timely Information for Practicing Physicians

October 1999

SPONTANEOUS BACTERIAL PERITONITIS (SBP)
Risk in cirrhotic patients. Mortality in SBP can be as high as 30%. Carlos Guarner and colleagues followed 109 cirrhotic patients with low ascitic fluid protein levels (< 1g/dL), which is known to confer an increased risk of SBP, to identify factors associated with a first episode of SBP (no patient received antibiotic prophylaxis). During a mean follow-up of 45 weeks, 28 patients developed SBP. A serum bilirubin level of >3.2 mg/dL and a platelet count <98,000/mm3 were associated with SBP (P<0.01 and <0.05, respectively). An accompanying editorial by Guadalupe Garcia-Tsao concluded that these two factors, along with low ascitic fluid protein levels, define patients at high-risk for SBP and is important for designing placebo-controlled prophylaxis trials. (Guarner C, et al. Gastroenterology 1999; 117: 414-419)
Effects of intravenous (IV) albumin. Pau Sort et al. randomized 126 cirrhotic patients with SBP to IV cefotaxime or IV cefotaxime plus IV albumin (1.5 g/kg at diagnosis and 1g/kg on day 3) to investigate if albumin therapy prevents SBP-related renal impairment. Renal impairment developed less frequently (P=0.002), and mortality rates during hospitalization and at 3 months were lower (P=0.01 and 0.03, respectively) in the albumin group. Nathan Bass emphasized in a related editorial that these results are encouraging and that the data suggest that the benefits of albumin therapy may be restricted to those with more advanced liver disease and/or severe renal impairment. (Sort P, et al. N Engl J Med 1999; 341: 403-409)

PRIMARY BILIARY CIRRHOSIS (PBC)
Epidemiology. Keiji Tsuji and coworkers reviewed the incidence of PBC in the Hiroshima prefecture (1988 to 1997) and found that 18 of 156 new patients were derived from 8 families in which at least one member had a history of PBC. The median age of PBC onset in this familial population was less than that in the general PBC population (33.4 ± 10.8 years vs. 55.6 ± 12 years). In a second study Oliver James et al. found the incidence and prevalence of PBC in the UK has risen since the 1980s and that mortality can be high even in PBC patients who present with asymptomatic disease. These studies suggest that a family history of PBC is a predisposing risk factor and that PBC is becoming a more commonly encountered liver disease. (Tsuji K, et al. J Autoimmunity 1999; 13: 171-178 and James OFW, et al. Hepatology 1999; 30: 390-394)
Effect of ursodeoxycholic acid (UDCA). Burton Combes and colleagues reviewed serial needle biopsy liver specimens from PBC patients in a 2-year, randomized, double-blind trial of UDCA (n=55) vs. placebo (n=60) to investigate the effect of UDCA on florid duct lesions. They found the prevalence of these lesions in the placebo and UDCA groups at both trial entry and at 2 years to be similar. These data indicate that - in contrast to its ability to palliate the effects of cholestasis - UDCA does not prevent bile duct destruction in PBC. (Combes B, et al. Hepatology 1999; 30: 602-605)
Low-dose methotrexate. Mark Hendrickse and associates conducted a 6-year, randomized, double-blind trial of low-dose methotrexate (n=30) vs. placebo (n=30) in PBC patients. The results showed that methotrexate treatment was associated with decreases in serum liver enzyme levels and immunoglobulin levels. However, deterioration of liver histology was not prevented. Moreover, there was a trend toward higher mortality and liver transplantation in the methotrexate group. In an accompanying editorial Paul Angulo and Rolland Dickson state that the discrepancy between the liver enzyme results and clinical outcomes is explained by the fact that these laboratory parameters are not predictive of PBC disease progression. They also question whether a larger dose of methotrexate might be more effective and assert that future trials must compare methotrexate with UDCA. (Hendrickse MT, et al. Gastroenterology 1999; 117: 400-407)

HEMOCHROMATOSIS
HFE gene expression. Two recent studies investigated the expression of the HFE gene. James Olynyk et al. studied 3,011 unrelated white adults in Brusselton, Australia. They found 0.5% of these adults (n=16) to be homozygous for the C282Y mutation. All homozygous subjects had elevated transferrin saturation (15 of 16 >45%), but only half of these subjects had clinical features of hemochromatosis and one-fourth had normal serum ferritin levels for up to 4 years of follow-up. In the second study, Antonello Pietrangelo et al. report on an Italian family in which 15 of 53 members had iron overload while none of the family members had the C282Y mutation. Anthony Tavill states in an editorial that the Australian study provides a model for genotypic-phenotypic correlation, with phenotypic expression guiding need for evaluation and treatment. He also comments that the results of the Italian study suggest there are other mutant genes that cause hemochromatosis. (Olynyk JK, et al. N Engl J Med 1999; 341: 718-724 and Pietrangelo A, et al. N Engl J Med 1999; 341: 725-732)