Hepatology highlights


  • Potential conflict of interest: Nothing to report.

Aggressive NASH is Strongly Associated With Adiponectin Deficiency

To more closely define the primary relevance of adiponectin to nonalcoholic fatty liver disease, Musso et al. studied patients with biopsy-proven nonalcoholic steatohepatitis (NASH) who were neither obese nor diabetic and compared their response to a fatty meal low in carbohydrates with that of 25 healthy controls matched for age, sex, body mass index, and waist-hip ratio. The four adipocyte-generated cytokines (adipokines)—tumor necrosis factor α, resistin, adiponectin, and leptin—were measured. The significantly reduced adiponectin levels found in patients with NASH were predictive of severe necroinflammation and advanced fibrosis on multiple ordinal regression analysis after controling for age, body mass index, waist circumference, insulin sensitivity index, and other cytokines (see Tables). Postprandial plasma levels of triglyceride and free fatty acids were significantly higher in patients with NASH than in controls and were inversely correlated with adiponectin levels. Not only were baseline serum adiponectin levels lower in subjects with NASH, but they also fell still further postprandially in contrast to the anticipated physiological rise seen in healthy controls. The data add to the mounting evidence that adipokines have a major influence on both the balance of lipid metabolism, which produces hepatic steatosis and the subsequent inflammatory and fibrotic changes that characterize NASH. (See HEPATOLOGY 2005;42:1175-1183.)1, 2

Table 1. Multiple Ordinal Regression Analysis for Factors Associated With Necroinflammatory Grade 3 in Subjects With NASH
FactorOdds Ratio95% CIP
Age (per 6-year increase)0.90.3–2.1.69
BMI (per 2 kg/m2increase)1.50.4–6.9.72
Waist (per 3-cm increase)0.80.3–1.9.59
ISI (per unit increase)1.90.6–5.9.23
Adiponectin (per 1,000-ng/mL decrease)5.01.1–28.1.009
Table 2. Multiple Ordinal Regression Analysis for Factors Associated With Fibrosis Stage 3 in Subjects With NASH
FactorOdds Ratio95% CIP
Age (per 6-year increase)0.50.1–1.9.27
BMI (per 2-kg/m2increase)4.10.4–40.1.20
Waist (per 3-cm increase)0.60.2–4.3.46
ISI (per unit increase)1.30.3–5.7.75
Adiponectin (per 1,000-ng/mL decrease)8.01.5–81.003

Choleresis Induced by Cholehepatic Cycling of Norurso (norUDCA)

Alan Hofmann's (AFH's) cholehepatic shunting bile acid analog does just what he always said it would. Two human volunteers were studied. (Was it fanciful on my part to assume that the initials AFH may have applied to a 65-year-old volunteer studied via a double lumen duodenal tube?) Natural human C24 bile acids have a 5-carbon (isopentanoic acid) side chain attached to a C19 steroid nucleus. NorC23 bile acids with a 4-carbon (isobutanoic acid) side chain occur as trace constituents in human bile but differ enormously from their C24 counterparts in most of their biological properties. Norursodeoxycholic acid (norUDCA) is readily protonated to its nonionic membrane permeable form, with the result that an estimated 97% was absorbed by proximal cholangiocytes, leaving its molar equivalent of bicarbonate ion in bile. The short-circuit rapid recycling of nor-bile acid in this way generates a vast bicarbonate-rich hypercholeresis, with an estimated trebling in daily volume of bile. Conjugation of norUDCA was mainly to its C23 ester glucuronide, which is freely excreted in urine, but N-acyl amidation by glycine and taurine, which is the norm for C24 bile acids, was seen to occur for norUDCA at only one sixteenth of the rate of glucuronidation. The concentrations of norUDCA in bile are substantially less than its critical micellar concentration; its resultant presence in monomeric rather than micellar form enhances its choleretic potential but diminishes biliary capacity for phospholipid and cholesterol secretion. More ingenuity, please, AFH, to suggest clinical exploitation of these impressive results. (See HEPATOLOGY 2005;42:1391-1398.)

URSO Outperforms Dexamethasone in Intrahepatic Cholestasis of Pregnancy

Intrahepatic cholestasis of pregnancy (IHCP) is not only associated with distressing pruritus in the mother, it may also result in spontaneous premature labor and carries an increased risk of intrauterine death for the fetus. Dexamethasone (DEXA) has gained a reputation for alleviating IHCP and is favored by obstetricians as a means of accelerating maturation of fetal lungs and thereby rendering premature labor, whether spontaneous or induced, less untimely for the infant. Glantz et al. report the first double-blind randomized controlled trial (DBRCT) comparing DEXA with ursodeoxycholic acid (URSO) in the treatment of IHCP. A total of 130 women with pregnancy-induced pruritus and serum bile acids >10 μmol/L were randomized to receive URSO, DEXA, or placebo. On an intention-to-treat analysis of the entire cohort, the only significant results related to falls in serum alanine aminotransferase (ALT) and bilirubin in the URSO group when compared to placebo. The same authors recently reported in HEPATOLOGY an observational study in a larger Swedish cohort of mothers with IHCP in which it was observed that morbidity was significantly greater in those with serum bile acid concentrations >40 μmol/L. When the 34 patients meeting this criterion were subjected to a subgroup analysis, treatment with both URSO and DEX significantly reduced serum bile acid levels compared with placebo, and reductions on URSO were significantly greater than with DEX. Reduced severity of pruritus occurred in response to URSO, being statistically superior to both DEX and placebo. Despite this confirmatory data that URSO provides symptomatic and biochemical benefit to affected mothers, the study lacked the power to demonstrate an effect of treatment on the most critical primary outcome variables of spontaneous preterm delivery and indicators of fetal distress. Nevertheless, I take the findings to be strongly supportive of recommendations for treatment of IHCP with URSO. Stronger evidence to justify treatment of IHCP with URSO would require both proof of improved fetal outcome and consent from a huge number of mothers with markers for the most severe end of the disease spectrum, to be randomized to placebo, almost certainly an unrealistic expectation given that in doing so they would be denying themselves an effective treatment for cholestasis, the source of both their personal distress and the threat to their fetus' wellbeing. (See HEPATOLOGY 2005;42:1399-1405.)

Female (Mainly) Traits in PBC

More than a thousand primary biliary cirrhosis (PBC) patients and controls matched for sex, age, race, and location completed telephone interviews of up to 180 minutes and comprising 180 questions and 300 subquestions. The results confirm that both genetic and environmental factors have a role in the pathogenesis of the disease. On multiple logistic regression analyses, PBC was significantly associated with a family history of PBC (adjusted OR [AOR], 10.736), systemic lupus erythematosus (AOR, 2.234) or Sjogren's syndrome (AOR, 5.814), an individual history of urinary tract infection (AOR, 1.511) and of smoking (AOR, 1.569), yearly use of nail polish (AOR, 1.002) and a history of the use of hormonal replacement therapy (AOR, 1.548). Never having been pregnant was associated with protection from developing PBC (AOR, 0.612). (See HEPATOLOGY 2005;42:1194-1202.)

Methotrexate Adds Nothing to URSO in Treatment of PBC

Treatment of PBC with methotrexate (MTX) has had its advocates, but they should now rest their case in light of the findings of a multicenter DBRCT reported by Combes and colleagues. Well-compensated patients with PBC who had taken URSO 15 mg/kg daily for at least 6 months were randomized to receive MTX 15 mg/m2 body surface area (n = 132) or placebo (n = 133) and followed for a median of 7.6 years. Liver biopsies taken before MTX and biannually thereafter showed a clear trend for worse outcome according to more advanced histological staging at entry, but despite having stratified patients accordingly and despite analysis according to myriad major and minor endpoints, the trial was brought to an early finish on the grounds of futility, as no advantage could be discerned from addition of MTX to URSO when compared with treatment with URSO alone (see Fig.). The merits of DBRCTs are thus unequivocal, but they represent an enormous challenge. Combes and colleagues calculate that in an 8-year study of moderately advanced PBC, an enrolment of 550 patients would be needed to demonstrate that a treatment halves the risk of requiring liver transplantation. (See HEPATOLOGY 2005;42:1184-1193.)

Illustration 1.

Detection and Correction of Vitamin A Deficiency in Cholestasis

Feranchak et al. reported on a battery of tests that have been proposed as measures of vitamin A deficiency. Vitamin A deficiency was detected in 43% of 23 children with cholestatic liver disease. Reduced baseline serum retinol concentrations of less than 20 μg/dL were found in 9 of 10 deficient individuals but also in 5 of 23 without deficiency (sensitivity 90%; specificity 78%). The gold standard was taken to be the intramuscular relative dose response (IMRDR), which was defined as the percentage rise in plasma retinol in response to an intramuscular injection of water miscible retinyl palmitate (1,500 μg). Plasma retinol does not rise appreciably (<10%) in those with sufficient stores of vitamin A, but a rise of more than 20% above baseline levels occurred in deficient individuals (see Fig.) The rise is attributed to a redistribution from the liver of the newly available vitamin A. In order to avoid any intramuscular injection, an oral RDR test was evaluated in which 450 μg of water-soluble retinyl palmitate is mixed with the water-soluble form of vitamin E, d-alpha tocopheryl polyethylene glycol-1000 succinate (TPGS). TPGS has an amphipathic structure known to enhance the absorption of lipid-soluble substances given concurrently. A rise of serum retinol level of more than 20% at 10 hours was detected in 8 of 10 vitamin A–deficient patients but was not seen in those with a false-positive serum retinol test (sensitivity 80%; specificity 100%). Malabsorption is the likely cause of a false-negative result. The authors recommended that serum retinol be used as a screening test and that in those with an abnormal result, the oral RDR be used for confirmation. Other indicators such as retinol binding protein (RBP), the retinol/RBP molar ratio, and various ophthalmic indications of deficiency, including conjunctival impression cytology, ophthalmologic slit lamp examination, Schirmer's test, and tearfilm break-up time were far less sensitive than the tests recommended above. (See HEPATOLOGY 2005;42:782-792.)

Illustration 2.