Potential conflict of interest: Nothing to report.
Adipokine Replacement Therapy Reverses NASH
Nonalcoholic steatohepatitis (NASH) is commonly a feature of the metabolic syndrome, characterized by insulin resistance, central obesity, and hypertension. The fat-laden splanchnic adipocyte secretes several proteins (adipokines) that influence hepatic glucose and lipid metabolism. Hepatic steatosis is promoted when secretion of tumor necrosis factor α is increased in proportion to its antagonists leptin and adiponectin, and recent evidence points to this imbalance as a mechanism for NASH. In severe lipodystrophy lack of adipose tissue is associated with low levels of these adipokines, and ectopic deposition of fat characteristically causes hepatomegaly. Javor et al. studied 10 patients with severe lipodystrophy and hypoleptinemia, 8 of whom had NASH confirmed histologically on baseline liver biopsy. Leptin replacement therapy was administered twice daily via subcutaneous injection of recombinant methionyl human leptin (R-metHuLeptin). Liver biopsies were taken before and after treatment for a median of 6.6 (range, 4-18) months. Repeat biopsies showed highly significant reductions in degree of steatosis and hepatocyte ballooning injury but no change in fibrosis. Magnetic resonance imaging confirmed that liver volume reduced by approximately 25% and liver fat content from 31% to 11%. Serum concentrations of triglyceride, insulin, glucose, and aminotransferase were significantly reduced. Lipodystrophy may be a rare specific cause of NASH, but the study confirms not only that impaired adipocyte function promotes ectopic fat storage but also that treatment of adipocyte endocrine dysfunction by replacement of deficient hormones (adipokines) in the tradition of Banting and Best may have a profound influence on this disease. Studies of leptin supplementation, while less intuitive, will be eagerly awaited in the large NASH population in whom leptin concentrations are not subnormal. Those patients, in stark contrast to lipodystrophy patients, are if anything overendowed with adipocytes. (See HEPATOLOGY 2005;41:753-760.)
Combined Urso and Budesonide in Treatment of PBC
Previous trials pointed to a beneficial effect of a combination of ursodeoxycholic acid (Urso) and corticosteroids on the natural history of primary biliary cirrhosis (PBC), but enthusiasm has been tempered with concern about the systemic side effects of steroid exposure. Budesonide has the apparent advantages of high first-pass extraction by the liver, higher affinity for glucocorticoid receptor binding, and conversion by the liver to relatively inactive metabolites, all of which combine to give it a far higher ratio of liver-bone glucocorticoid exposure in comparison with prednisolone. These apparent advantages are negated when first-pass extraction is impaired; recent studies published in HEPATOLOGY have shown that increased systemic exposure to budesonide in patients with cirrhosis and portal hypertension caused unacceptable side effects such that it is now contraindicated in patients with stage IV PBC and/or porto-systemic shunting. The study reported by Rautainen et al. is timely given that it prospectively excluded patients with those features of advanced disease and used a lower dose of budesonide (6 mg/day) compared to previously reported studies in which patients generally received 9 mg/day. Seventy-seven patients with PBC were randomized to receive Urso (15 mg/kg/day) plus budesonide or Urso alone. Liver biopsies taken before and after 36 months of treatment in 69 patients revealed significant differences in favor of the combination treatment for change of stage (P = .009), with a 22% improvement compared with a 20% deterioration on Urso alone. Similarly, fibrosis decreased 25% in the budesonide-treated group but increased 70% on Urso alone (see Fig.). The failure of inflammation to show the statistically significant improvement reported in previous trials of budesonide in PBC may be a result of the reduced dose. Improvement in stage was significantly greater in the combination arm only for those with stage II disease at entry, supporting the proposal that efficacy is more readily demonstrable at earlier stages of the disease. Although serum bilirubin, the most prognostic biochemical measure in PBC, remained within the normal range, it rose significantly on Urso alone but remained stable in those on budesonide and Urso in combination. Two patients required dose reduction or cessation of budesonide because of systemic glucocorticoid exposure. (See HEPATOLOGY 2005;41:747-752.)
Long-Term Outcome of OLT for HPS
Hepatopulmonary syndrome (HPS), a rare complication of chronic liver disease, causes severe hypoxemia due to dilatation of the pulmonary microcirculation. Swanson et al. reviewed the clinical course of 61 patients with HPS who were seen at the Mayo Clinic between 1985 and 2002. In 14 patients awaiting orthotopic liver transplantation (OLT), sitting PaO2 measurements (on room air) were observed to decline progressively by a mean of 5.2 ± 2.3 mmHg per year (median, 6.3; range, 0.4-8.3). Of the entire cohort of HPS patients, those who underwent transplantation had superior survival when compared to the remainder who had either died awaiting transplant or who were not selected for OLT because of co-morbidity. Comparison of the nontransplanted HPS patients with a case control group matched for diagnosis, age, and severity of liver failure demonstrated markedly inferior median survival (24 vs. 87 months) and 5-year survival rates (23% vs. 63%, P = .0003). Five-year survival following liver transplantation was comparable in those with HPS and controls (76% vs. 84%). Outcome of OLT was inferior in HPS patients with PaO2 of < 50 mmHg (see Fig.). Swanson et al. commented that in the early years of their program HPS was regarded as a relative contraindication to OLT but that HPS in the absence of comorbities, regardless of its severity, should no longer lead to exclusion of a transplant candidate. They argue plausibly that HPS should contribute to the Model for End-Stage Liver Disease (MELD) score and that progression of hypoxemia beyond 60 toward 50 mmHg should be a signal for granting the recipient high priority. (See HEPATOLOGY 2005;41:1122-1129.)
HRS and MELD–A Bridge Too Far for OLT
Prognostication in end-stage liver disease has assumed increasing importance since the advent of liver transplantation. The MELD score is now accepted as a fair means of giving priority when allocating donor livers. Candidates for transplant need advice that permits them the optimum chance not only of receiving a graft but also of doing so before their condition has deteriorated to a state whereby the chances of survival have been diminished. In virtually all prognostic models for posttransplantation mortality, serum creatinine has proven to be one of the strongest predictors. In hepatorenal syndrome (HRS), liver failure is associated with functional renal failure in the absence of intrinsic renal disease. The report by Alessandria et al. analyzed the survival of 105 consecutive HRS patients admitted to the Barcelona unit between 1992 and 2001. Patients were classified according to the 1996 definitions of the International Ascites Club into type 1, which is characterized by rapidly progressive renal failure, and type 2, in which renal failure is more moderate and stable. On multivariate analysis only the MELD score and type of HRS were associated with independent prognostic value. In type 1 HRS median survival was 1 month almost regardless of MELD scores ranging from 20 to 45. In type 2 HRS median survival was 6.7 months but varied significantly according to MELD score. For type 2 HRS patients with a MELD score greater than 20, survival was 1 month, compared with 8 months for those with a MELD score lower than 20. The MELD score proved superior to the serum creatinine (a key component of MELD) in predicting 3-month mortality using ROC curve analysis (P < .05). Many of the patients in this series participated in well-designed studies on the pathophysiology of HRS and trials of treatment that included infusion of intravenous vasopressor agents and transjugular portosystemic shunts, which sadly did not have a statistically significant effect on survival. The extreme gravity of developing HRS is well illustrated by this report and that clinical message hardly needs reinforcement by MELD. The corollary may be the key message: those with a MELD score of less than 20 have a seriously worse prognosis than others with an equivalent score once HRS has developed. (See HEPATOLOGY 2005;41:1282-1289.)
Comeback Time for Gallstones
Mature readers of HEPATOLOGY will remember vintage AASLD meetings when the latest buzz was bile and lithogenic indices were all the rage. Those nostalgic for those days will have appreciated two recent studies that analyzed the potential of bile to form microcrystals in relation to bariatric surgery and biliary pancreatitis, respectively.
Gustafsson et al. aspirated bile from the gallbladder during gastroplasty for morbid obesity and by percutaneous gallbladder puncture 1.1 to 7.3 months later in 25 subjects. Gallstones were not present initially but developed in 2 subjects during follow-up. In bile taken less than 2 months following gastroplasty (13 patients) there was a constellation of findings that would support gallstone formation. Cholesterol saturation was increased (see Fig.) and crystal detection time was shortened from 24 to 7 days. Indirect evidence pointed to increased presence of both a protein (conconavalin-A–Sepharose extractable) and mucin (detected as hexosamine) with crystallization-promoting properties. These factors were all enhanced during the early postoperative phase, which was associated with rapid weight loss, but had largely returned to baseline levels in gallbladder bile aspirated 2 to 8 months after bariatric surgery. Increasing focus on weight loss as therapy in NASH should not obscure a possible need for concomitant gallstone prevention. (See HEPATOLOGY 2005;41:1322-1328.)
Venneman et al. studied gallbladder bile obtained at cholecystectomy and found highly significant differences between those with and those without a prior history of pancreatitis. Compared with those having cholecystectomy for symptomatic but uncomplicated gallstones, those with a history of pancreatitis had more strongly contracting gallbladders, despite similar plasma cholecystokinin profiles, higher prevalence of sludge, and smaller and more numerous gallstones. Crystallization occurred significantly faster in gallbladder bile from pancreatitis sufferers, despite similar content of bile acids, biliary lipid, and pronucleating proteins. Concentrations of gallbladder mucin were higher in the pancreatitis group and are likely to have been the critical factor that promoted faster crystallization and more sludge formation in patients with pancreatitis. (See HEPATOLOGY 2005;41:738-746.)