Hepatology highlights


  • Potential conflict of interest: Nothing to report.

The Failing Heart and HRS

The contribution of suppressed cardiac function to the hepatorenal syndrome (HRS) is emphasized in a highly informative prospective study reported by Ruiz-del-Arbol et al. Haemodynamic studies were performed at baseline in 66 patients with cirrhosis and tense ascites. They were repeated under the same standardized conditions in 27 who developed the hepatorenal syndrome during the subsequent 2 years. As anticipated, those who developed hepatorenal syndrome-type 2 (characterized by moderate, relatively stable renal failure) had, at baseline, lower mean arterial pressure, higher plasma renin activity and greater sympathetic nervous system activity. The HRS group had higher wedged hepatic venous pressure and wedged hepatic venous pressure gradients at baseline; on development of HRS portal pressure rose even further. Total systemic vascular resistance was surprisingly not different between those who would or would not subsequently develop HRS. The latter implies that the systemic vasoconstrictor responses in those who progressed to type 2 HRS were successfully compensating for splanchnic vasodilatation at that stage. More surprisingly, those with the more advanced chronic liver disease who went on to develop HRS already had significantly lower cardiac outputs at baseline. This finding subverts the traditional notion that cirrhosis at this stage is associated with an unrestricted ability of the heart to increase cardiac output to maintain hemodynamic stability and that the critical change which precipitates HRS is progressive splanchnic vasodilatation. A lower cardiac output and increased plasma renin activity were the only two of ten variables measured at baseline which were independently predictive of HRS (Fig.). In the presence of HRS, diminished hepatic and renal blood flow were associated with a further fall in cardiac output and a failure to increase heart rate despite the obvious physiological demand. Cardiopulmonary studies demonstrated an association of Type 2 HRS with a trend towards lower cardiac output, stroke volume, stroke work, pulmonary capillary wedge pressure, and right atrial pressure. These findings point to a reduction of cardiac preload as an important mechanism of the fall in cardiac output. The information on diminished preload provides a satisfying mechanistic explanation for the proven efficacy of plasma expansion with colloid infusion in prevention of type 2 HRS. The popular wisdom regarding splanchnic vasodilatation and activation of renal vasoconstriction as mechanisms producing HRS would on this evidence need to be modified to incorporate a role for a heart which is refractory to demands for increased hepatic blood flow and concedes attempts to maintain mean arterial pressure to vasoconstrictor mechanisms which, by default of the heart and at their most exuberant, precipitate HRS. (See HEPATOLOGY 2005;42:439–447.)

Illustration 1.

Quality of Life in TIPS vs. Paracentesis for Refractory Ascites

The choice of transjugular intrahepatic portal-systemic shunting (TIPS) and repeated paracentesis for refractory ascites would each have supporters and opponents on the basis of clinical experience. Campbell and colleagues report the outcome of these procedures in 109 patients randomly allocated to either procedure in an international prospective controlled trial. A quality-of-life analysis was performed by SF-36 survey at baseline and after 6 and 12 months' follow-up. Because of death and transplantation, only 51 patients completed the 12 months' assessment by which time those in the TIPS group could be shown to have experienced less number of large volume paracentesis, lower diuretic dosages, less shortness of breath and abdominal distension, but higher rates of confusion and hospitalisations. When the physical component score (PCS) of SF-36 was analysed by multivariate analysis, no clear advantage emerged in favour of either therapeutic approach though significant improvement was associated with lack of confusion, improved ascites and lack of hospitalisations. Improvement in the mental component scale of SF-36 was associated with lack of confusion and randomization to TIPS. Avoidance of encephalopathy, therefore, stands out as a critical endpoint in management of decompensated chronic liver disease. When treating refractory ascites the balance tips in favour of TIPS when it is not complicated by encephalopathy. Such data emphasize the vital importance of patient selection for TIPS, with particular focus on the risk of encephalopathy following the procedure. (See HEPATOLOGY 2005;42:635–640.)

Albumin vs. Starch With Paracentesis

The Barcelona group continue their logical progression through studies on clinical management of the complications of cirrhosis. They had previously reported improved survival due to albumin infusion upon first diagnosis of spontaneous bacterial peritonitis (SBP) and that hydroxyethyl starch 200/0.5 (HES) is as effective as albumin in the prevention of circulatory problems following large volume paracentesis. Fernández et al. report on a pilot study in patients with spontaneous bacterial peritonitis. Twenty patients with cirrhosis and SBP were randomized to receive either albumin or HES. Albumin protected significantly against circulatory failure with statistically significant changes in numerous variables, including increases of mean arterial pressure and systemic vascular resistance and reductions in heart rate and plasma renin activity. However, HES did not result in a statistically significant change in any of these measurements of circulatory function. Measures of endothelial dysfunction, including plasma levels of von Willebrand-related antigen, nitrates and nitrites, were increased on HES but not on albumin (Fig.). Four of ten patients treated with HES developed SBP-induced circulatory dysfunction or renal failure compared with none of ten infused with albumin. Although the oncotic capacity of equivalent weights of albumin and HES are identical, their half-life in plasma and the total volume of infusate differ markedly. Furthermore, HES lacks the beneficial effects of albumin in countering the systemic inflammatory response and endothelial dysfunction which escalates harm from SBP. (See HEPATOLOGY 2005;42:627–634.)

Illustration 2.

Detecting Cancer in Cirrhotic Nodules

The detection of nodules in the liver with cirrhosis raises the critically important question as to whether malignancy is present or not. Guidelines published by the European Association for Study of the Liver state that arterial hypervascularity on two different imaging techniques in nodules greater than 2 cm in diameter is sufficient evidence for the diagnosis of hepatocellular carcinoma (HCC). Bolondi et al. prospectively evaluated these guidelines with contrast perfusional ultrasonography and helical computed tomography of seventy two nodules in 59 patients who were examined prospectively and extended their study to include nodules less than 2 cm. Biopsy was performed when one or both imaging techniques did not show hypervascularity and was repeated after 3 months if histology was negative for HCC but the lesion had enlarged significantly. Hypervascularity was demonstrated by both techniques in 44% of 1-2 cm nodules and in 84% of 2–3 cm nodules. HCC was found in 4 of 14 nodules which were hypovascular by both techniques and in 11 of 14 nodules which were hypervascular by just one. All nodules 2.1 – 3 cm in diameter eventually proved to be HCC. With respect to routine surveillance of cirrhotic populations, the study begs the question of whether all nodules need to be biopsied to prove the absence of HCC or whether close surveillance with attention to progressive enlargement should be a prerequisite of biopsy in hypovascular nodules less than 2 cm in diameter. Concerns regarding cancer seeding and other complications of needle biopsy have to be weighed against an as yet unproven benefit from early diagnosis.(See HEPATOLOGY 2005;42:27–34.)

Proteomics and Genomics in NASH

All hypotheses and theories are set aside when gene expression in the liver and plasma protein profiles are examined by quantification of messenger RNA from genes known to be expressed in the liver and plasma proteins quantified by SELDI-TOF mass spectrometry. Younossi et al. processed liver biopsies taken from obese patients at the time of bariatric surgery which were classified as normal (7), steatosis alone (12) steatosis with nonspecific inflammation (52) and nonalcoholic steatohepatitis (NASH) (27). Using a threshold of a greater than twofold difference in hepatic gene expression and in comparison with non steatotic biopsies from obese controls 7 genes were differentially expressed in steatosis alone and 14 in NASH. Downregulation of phase II detoxifying mechanisms was apparent in steatosis alone for Mu-class glutathione sulphotransferase and in later stages of NAFLD for cytosolic sulfotransferase isoform 1A2. Upregulated genes included those related to stellate cell activation and fibrogenesis and a shift in cellular homeostasis towards apoptosis. Comparisons between NAFLD groups revealed only two greater than twofold differences; compared to steatosis alone, in NASH expression of IGFBP1 was significantly downregulated and of FACL4 upregulated. Twelve plasma protein peaks were expressed significantly differently between NAFLD and obese controls. As cogniscenti mull over the finer detail revealed by data generated in this and similar studies we trust they will be inspired to generate insightful hypotheses to direct future research endeavour. (See HEPATOLOGY 2005;42:665–674.)

NAFLD in Two Guises and in Defiance of Simple Identification Methods

Attempts to define a simple screening tool which identifies NAFLD in large population studies remain frustratingly unsuccessful despiteseveral recent studies which have attempted to refine the clinical and epidemiological criteria which characterize NAFLD and nonalcoholic steatohepatitis (NASH). The Dionysos project invited all residents from two towns to participate in a study of chronic liver disease in 1991 and a decade later. When subjects with elevated serum alanine aminotransferase (ALT) and γ-glutamyl transferase were compared with randomly selected controls matched for age and sex there was no difference in the incidence (25% and 20%, respectively) of fatty liver diagnosed on ultrasonographic criteria. (Bedogni et al.).1 Concerned by such lack of sensitivity, and in order to increase detection of NAFLD by means of screening blood tests, Kunde et al.2 proposed a new normal range for ALT in obese women. By lowering the upper limit of normal (ULN) for ALT from <30 U/L to <19 U/L the proportion of women with class II/III obesity who had normal ALT was reduced from 72.5% to 36.9% but the heightened sensitivity failed to increase the specificity for fatty liver, portal fibrosis with steatosis and NASH. Those declared abnormal according to the ALT criterion only following reduction of the ULN comprised a significantly greater proportion with steatosis alone but a decreased prevalence of NASH compared to those with ALT >30. Reducing the ULN from 30 to 19 increased the sensitivity for NASH from 42% to 74% but simultaneously reduced specificity from 80% to 42%.

The importance of portal fibrosis with steatosis as a variant of NAFLD was further emphasized by a study in 100 children aged 2 to 18 years who had undergone liver biopsy (Schwimmer et al.).3 Histological features of steatohepatitis fell into 2 distinct patterns by agglomerative hierarchical cluster analysis. The classical picture of NASH with ballooning degeneration and pericellular fibrosis (type 1) was seen in 17% but in 51% of the children NASH was characterized by portal inflammation and portal fibrosis (Type 2) (fig.). Type 2 NASH was significantly associated with male sex, non-white race, and more severe stages of fibrosis. The natural history of NASH may differ significantly according to these histological subtypes.

Illustration 3.