In accordance with increasing demands for evidence-based practice for ethical, fiscal, and medico-legal reasons HEPATOLOGY provides a rich source of material that can justly guide the hepatologist's day-to-day clinical activity. The first four articles highlighted below constitute such confidence building data.
Provide feedback or get help You are viewing our new enhanced HTML article.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Should Paracentesis be Covered With Infusion of Platelets and Plasma?
Many patients with cirrhosis who have diuretic-resistant ascites will have abnormally low platelet counts and prolonged prothrombin times, which, in a climate of defensive medicine, may promote anxiety about the safety of large-volume paracentesis in the absence of prior infusion of platelets and fresh-frozen plasma. Current guidelines from the American Association for the Study of Liver Diseases state that such hematological support is not routinely indicated, and this guideline is supported in the study of Grabau et al. Gastrointestinal endoscopy assistants who had been trained in the technique performed 1,100 large-volume paracenteses in 628 patients without any procedure-related complication. Preprocedure, the mean international normalized ratio was 1.7, and the mean platelet was count 50.4 × 10 3/μL. For purposes of training, the authors recommend that operators should perform an initial 10 paracenteses under supervision. (See HEPATOLOGY 2004;40:484–488.)
TIPS Versus Paracentesis For Ascites
When ascites is resistant to medical management, the balance of risk and benefit from treating it by reducing portal pressure via portosystemic shunting is extremely difficult to judge. Reduced hepatic perfusion resulting directly from lowered perfusion pressure may precipitate hepatocyte failure, which in turn compounds the risk of encephalopathy from increased shunting. Salerno et al. report the results of a trial in which patients with refractory or recidivant ascites were prospectively randomized to transjugular intrahepatic portosystemic shunting (TIPS) or repeated paracentesis with albumin infusion. Previous trials comparing these modalities have provided conflicting outcomes. Survival at 1 and 2 years was significantly better in those randomized to TIPS (Fig. 1). Although more episodes of severe (grades 3 and 4) encephalopathy occurred in the TIPS group, the outcomes on balance justify choice of TIPS in selected cases for treatment of diuretic-resistant ascites; the key aspect remains the detail of patient selection. Although improved survival could be shown for Child-Turcotte-Pugh Class C patients, it is important to note that anyone with a Child-Turcotte-Pugh score higher than 11 was excluded. The results would certainly appear to favor TIPS over paracentesis for management of resistant or recalcitrant ascites in appropriately selected candidates for whom subsequent liver transplantation remains an option. (See HEPATOLOGY 2004;40:629–635.)
Can Diagnostic ERCP Still be Justified for PSC?
In this study, the cost and diagnostic efficiency of endoscopic retrograde cholangiopancreatography (ERCP) and magnetic resonance cholangiography (MRC) were compared for the initial diagnosis of primary sclerosing cholangitis. The results should signal a uniform change in practice that has its base as much in the ethical and medico-legal arenas as in the fiscal arena. Seventy-three patients suspected of having biliary disease underwent ERCP within 24 hours after MRC. In computing the cost-effectiveness of the approaches, a significant sum was allocated for management of post-ERCP complications. On cost and efficiency criteria alone, the authors state that MRC is at least comparable to ERCP; therefore, given the possibly life-threatening complications of ERCP and the relative safety of MRC, it is open to debate whether future guidelines incorporate a view that ERCP is not indicated for diagnosis of PSC even in the face of a normal MRC, in view of the risks involved and the relative lack of therapeutic and prognostic implications of making an earlier diagnosis. (See HEPATOLOGY 2004;40:39–45.)
Two articles on intrahepatic cholestasis illustrate the equally valid but complementary approaches of laboratory science and clinical observation to our knowledge base.
The Importance of Serum Bile Acids in Pregnancy
Intrahepatic cholestasis of pregnancy (ICP) is known to cause distressing pruritus in the mother and to threaten the health and viability of the fetus, but identifying a specific risk factor and quantifying the risk has not previously been achieved. In ahighly informative prospective study from Sweden, Glantz et al. screened 937 women who developed pruritus of pregnancy that had no dermatological explanation. Diagnostic criteria for ICP also included having serum bile acid concentrations of 10 μmol/L or greater and were fulfilled in 693 patients, constituting an incidence for ICP of 1.5% in all pregnancies, and excluded 26% of those presenting with pruritus of pregnancy. There was a weak but significant correlation between bile acid concentrations and the severity of pruritus recorded on a visual analogue scale, but a poor correlation with alanine aminotransferase, strengthening the case for regarding measurement of serum bile acid as the gold standard in diagnosis and monitoring of ICP. Correlations were found between elevated serum bile acid levels and fetal complication rates with increased incidence of spontaneous preterm delivery, asphyxial events, meconium staining of liquor, and green staining of placenta and membranes. Additional statistical analysis by spline functions stratified ICP subjects into a low-risk group (81% of the cohort) and a high-risk group (19%) in whom serum bile acids were 10–39 and >40 μmol/L, respectively (Fig. 2). Complications did not occur with enhanced frequency in ICP when compared with ICP pregnancies below a threshold bile acid concentration of 40 μmol/L but above this increased by 1% ± 2% for each additional 1 μmol/L rise in bile acid concentration. In those with a diagnosis of severe ICP, there was a more frequent history of intrauterine fetal death (IUFD) in a previous pregnancy (4.1% compared to 0.4% of controls), preterm delivery (25% vs. 4%), gallstone disease (7.4% vs. 0.5%), and heredity for pruritus of pregnancy (30% vs. 13%). Active management of the pregnancies probably resulted in a very low rate of IUFD during the study. Although not reaching statistical significance, there was a counterintuitive trend for onset of pruritus to occur later in pregnancy with severe ICP. If confirmed, these results could significantly influence the future management of ICP. The weekly measurements of serum bile acids may be unnecessarily frequent, but the broad strategy of monitoring serum bile acid levels in expectant mothers who complain of pruritus to identify a subgroup (19% in this population) in whom active management minimizes risk of fetal distress and IUFD is compelling. (See HEPATOLOGY 2004;40:467–474.)
Surprisingly Diverse Phenotypic Expression of ATP8B1 Mutations
This detailed study of the genetic abnormalities found in 180 families with progressive familial intrahepatic cholestasis (PFIC) or benign recurrent intrahepatic cholestasis (BRIC) further illustrates the complex relationship that exists between the presence of a given mutation and disease expression. Given that BRIC is a remitting disease that leaves no lasting liver injury and that PFIC leads to progressive fibrosis and end-stage liver disease, it may seem surprising that the same gene, ATP8B1, which encodes for the ABC bile canalicular membrane transporter FIC1, may be at fault. Moreover, although there was a tendency for single-point mutations to be more commonly associated with BRIC and for more gross mutations such as large gene deletions to be associated with PFIC, this pattern was not consistent (Fig. 3). ATP8B1 mutations were identified in 30% of the 130 PFIC families tested and in 40% of BRIC families. The thoroughness of the search is reflected by detection of mutation on both alleles in 92% of patients in whom any mutation was found. Therefore, few of the remaining 70% of PFIC patients and 83% of BRIC patients not carrying the common I661T mutation are likely to have as yet undetected mutations affecting FIC1. A similarly thorough exclusion of mutations in the ABCB11 gene, which codes for the ABC canalicular bile salt transporter BSEP, is predicted to define a significantly large proportion of patients with syndromic BRIC and PFIC who characteristically have normal serum levels of in whom candidate genes other than those encoding FIC1 and γGT BSEP are likely to be incriminated. (See HEPATOLOGY 2004;40:27–38.)
Adipokine Imbalance and NASH
The adipocyte, now known to be much more than a fat storage cell, secretes several metabolically active proteins collectively known as adipokines, which influence hepatic glucose and lipid metabolism. One of these proteins, tumor necrosis factor α (TNFα) has been widely implicated in the pathogenesis of nonalcoholic steatohepatitis (NASH), as it induces insulin resistance and is a proinflammatory cytokine. Other adipokines include leptin and adiponectin, the latter in contradistinction to TNFα possessing antilipogenic and anti-inflammatory properties. Hui et al. report on a study in 109 patients with steatosis, including 80 with NASH and 82 matched volunteers. Serum adiponectin levels were significantly different among the 3 subgroups being lowest in NASH and intermediate in simple steatosis (Fig. 4). As has been previously described, the other adipokines measured (leptin and TNFα) as well as TNF receptor 2 (TNFR2), a marker of activation of the TNFα system that correlates well with insulin resistance) all showed a trend to increasing serum levels in steatosis and NASH with the highest levels being found in NASH, which is the inverse of the trend found for adiponectin. In multiple logistic regression analysis reduced adiponectin, increased TNFα and TNFR2 were independently associated with NASH when compared with controls; but when NASH and simple steatosis groups were compared, only increased insulin resistance by HOMA-IR (odds ratio, 1.7; CI, 1.3-2.3; P = .001) and reduced adiponectin levels (odds ratio, 1.7; CI 1.0-3.0; P = .06) showed a clear trend to being independent predictors of NASH. The receiver operating characteristic curve (Fig. 5) identified an adiponectin level less than 10 μg/mL and a HOMA-IR >3 units as identifying 77% of NASH subjects but only 33% of those with simple steatosis. After controlling for HOMA-IR, decreased serum adiponectin levels were associated with greater severity of steatohepatitis on liver biopsy, such that values <10 μg/mL were found in 84% of those with necroinflammatory grades 2 and 3 but in 52% of those with simple steatosis. These results support the notion that reduced adiponectin levels are involved in the transition from simple steatosis of the liver to NASH. In discussion of their findings, Hui et al. cite clinical and experimental data from elsewhere that is supportive of this proposal. (See HEPATOLOGY 2004;40:46–54.)