Hypothalamus-pituitary dysfunction is common in patients with stable cirrhosis and abnormal low dose synacthen test.

BACKGROUND: Adrenal insufficiency is often present in cirrhosis. We hypothesize that a prolonged adrenocorticotropic hormone (ACTH) stimulus can restore cellular capacity of adrenal glands to secrete cortisol. Aim of our study was to assess adrenal responsiveness to prolonged ACTH stimulation in cirrhotics. METHODS: Prospective observational study in 121 consecutively admitted cirrhotic patients undergoing a low dose short synacthen test and plasma ACTH measurement using a chemiluminescence immunoassay. Long synacthen test was performed if the low dose was abnormal. RESULTS: 46 patients had abnormal low dose short test (38%), and 29 underwent the long test: 41% showed normal response (Group 1), 55% showed delayed response (Group 2) and 1 had abnormal response (4%). Baseline ACTH levels did not significantly differ between the two groups. Median basal cortisol was higher in Group 1 (296 vs. 198 nmol/L; p=0.02). Using ROC curve basal cortisol <254 nmol/L was associated with a delayed long synacthen test response (AUC 0.78, p=0.001) with good accuracy (sensitivity 67%, specificity 81%). CONCLUSION: A delayed cortisol response after a prolonged ACTH stimulation is found in over fifty percent of cirrhotics with abnormal low dose short synacthen test, confirming that the mechanism of hypoadrenalism in these patients could be related both to adrenal cellular dysfunction and hypothalamus-pituitary adrenal axis impairment.

Anti-tumour necrosis factor agent and liver injury: literature review, recommendations for management.

Abnormalities in liver function tests, including transient and self-limiting hypertransaminasemia, cholestatic disease and hepatitis, can develop during treatment with anti-tumour-necrosis-factor (TNF) therapy. The optimal management of liver injury related to anti-TNF therapy is still a matter of debate. Although some authors recommend discontinuing treatment in case of both a rise of alanine aminotransferase more than 5 times the upper limit of normal, or the occurrence of jaundice, there are no standard guidelines for the management of anti-TNF-related liver injury. Bibliographical searches were performed in PubMed, using the following key words: inflammatory bowel disease (IBD); TNF inhibitors; hypertransaminasemia; drug-related liver injury; infliximab. According to published data, elevation of transaminases in patients with IBD treated with anti-TNF is a common finding, but resolution appears to be the usual outcome. Anti-TNF agents seem to be safe with a low risk of causing severe drug-related liver injury. According to our centre experience, we found that hypertransaminasemia was a common, mainly self-limiting finding in our IBD cohort and was not correlated to infliximab treatment on both univariate and multivariate analyses. An algorithm for the management of liver impairment occurring during anti-TNF treatment is also proposed and this highlights the need of a multidisciplinary approach and suggests liver biopsy as a key-point in the management decision in case of severe rise of transaminases. However, hepatic injury is generally self-limiting and drug withdrawal seems to be an exception.

Liver transplantation for unresectable neuroendocrine tumor liver metastases.

BACKGROUND: Liver transplantation (LT) is performed in selected patients with neuroendocrine hepatic metastases. Survival benefit and the risk of tumor recurrence after LT, also exacerbated by immunosuppressive therapy, remain important clinical issues. Whether patients with particular types of neuroendocrine tumors (NET) benefit more than others is unclear. METHODS: Bibliographical searches were performed in PubMed for the terms "liver transplantation and neuroendocrine tumors," "liver transplant and neuroendocrine tumors," "liver transplantation and immunosuppressive therapy," "tumor recurrence." RESULTS: Promising results have been reported for LT for NET metastases with 5-year survival of up to 90 % in patients with well-differentiated gastroenteropancreatic NETs, but only few patients are free of tumor 5 years after LT. Better outcomes have been reported for gastrointestinal tumors than for pancreatic NETs for both survival and risk or recurrence after LT. Selection criteria for LT are limited and include the 2007 Milan Criteria and the 2012 European Neuroendocrine Tumor Society guidelines, including: well-differentiated NET (Ki-67 <10 %), age <55 years, absence of extrahepatic disease, primary tumor removed before transplantation, stable disease for at least 6 months before LT, and <50 % liver involvement. CONCLUSIONS: LT might be considered in carefully selected patients. The risk of tumor recurrence remains a significant clinical problem after LT, but data focused on immunosuppression issue are lacking, and there are no currently approved strategies for prevention of recurrence or follow-up protocols. Further studies are needed to define universally accepted inclusion criteria, reliable predictors of better outcome, and optimal timing for LT.

Comparison of total cortisol, free cortisol, and surrogate markers of free cortisol in diagnosis of adrenal insufficiency in patients with stable cirrhosis.

BACKGROUND & AIMS: Measurements of serum levels of total cortisol can overestimate the prevalence of adrenal dysfunction in patients with cirrhosis because they have low concentrations of corticosteroid-binding globulin and albumin. We used measurements of serum total cortisol and serum free cortisol after the low-dose short Synacthen test (LDSST) to assess adrenal dysfunction. METHODS: We studied 79 patients with stable cirrhosis; adrenal dysfunction was defined by peak concentrations of total cortisol ≤494 mmol/L and/or peak concentrations of free cortisol ≤33 nmol/L after the LDSST. We determined free cortisol index (FCI) scores and calculated free cortisol levels by using Coolens' equation. The Cox regression model was used to assess the relationship between adrenal dysfunction and outcomes (death or liver transplant). RESULTS: On the basis of measurement of total cortisol, 34% of patients had adrenal dysfunction, and on the basis of measurement of free cortisol, 29% had adrenal dysfunction. There was agreement between total cortisol and free cortisol levels in 22% of patients; in 13%, adrenal dysfunction was diagnosed only on the basis of total cortisol and in 6% only on the basis of free cortisol (κ coefficient, 0.56; P < .01). Low concentrations of corticosteroid-binding globulin (21 vs 54 µg/mL, P < .01) led to an overestimation of adrenal dysfunction that was based on measurement of total cortisol. Measurements of calculated free cortisol constantly overestimated free cortisol concentrations, with variations as large as 87% for baseline values and up to 84% after stimulation. Adrenal insufficiency, defined by FCI scores <12, was detected in 30% of patients; among them, 23% also had subnormal peak levels of free cortisol (κ coefficient, 0.70; P < .001). Adrenal dysfunction was not significantly associated with patient outcomes, on the basis of Cox model analysis. CONCLUSIONS: Adrenal insufficiency, defined by LDSST, is frequent in patients with stable cirrhosis, on the basis of measurements of total and free cortisol. FCI scores are better than measurement of total cortisol in assessing adrenal function in patients with cirrhosis. We did not associate adrenal dysfunction with outcome, but further studies are needed.

Reduced exposure to calcineurin inhibitors early after liver transplantation prevents recurrence of hepatocellular carcinoma.

BACKGROUND & AIMS: Recurrence of hepatocellular carcinoma (HCC) is a major complication after liver transplantation (LT). The initial immunosuppression protocol may influence HCC recurrence, but the optimal regimen is still unknown. METHODS: 219 HCC consecutive patients under Milan criteria, who received an LT at 2 European centres between 2000 and 2010, were included. Median follow-up was 51 months (IQR 26-93). Demographic characteristics, HCC features, and immunosuppression protocol within the first month after LT were evaluated against HCC recurrence by using Cox regression. RESULTS: In the explanted liver, 110 patients (50%) had multinodular HCC, and largest nodule diameter was 3±2.1cm. Macrovascular invasion was incidentally detected in 11 patients (5%), and microvascular invasion was present in 41 patients (18.7%). HCC recurrence rates were 13.3% at 3 years and 17.6% at 5 years. HCC recurrence was not influenced by the use/non-use of steroids and antimetabolites (p=0.69 and p=0.70 respectively), and was similar with tacrolimus or cyclosporine (p=0.25). Higher exposure to calcineurin inhibitors within the first month after LT (mean tacrolimus trough concentrations >10ng/ml or cyclosporine trough concentrations >300ng/ml), but not thereafter, was associated with increased risk of HCC recurrence (27.7% vs. 14.7% at 5 years; p=0.007). The independent predictors of HCC recurrence by multivariate analysis were: high exposure to calcineurin inhibitors defined as above (RR=2.82; p=0.005), diameter of the largest nodule (RR=1.31; p<0.001), microvascular invasion (RR=2.98; p=0.003) and macrovascular invasion (RR=4.57; p=0.003). CONCLUSIONS: Immunosuppression protocols with early CNI minimization should be preferred in LT patients with HCC in order to minimize tumour recurrence.

Early tacrolimus exposure after liver transplantation: relationship with moderate/severe acute rejection and long-term outcome.

BACKGROUND & AIMS: Liver transplant (LT) patients might be overimmunosuppressed as recommendations for tacrolimus trough concentrations (TC) within 4-6 weeks after liver transplantation are set too high (10-15 ng/ml). Early tacrolimus exposure was evaluated in relation to acute rejection and long-term outcomes. METHODS: Four hundred and ninety-three consecutive LT patients receiving tacrolimus as primary immunosuppression (1995-2008) were analyzed. Acute rejection was diagnosed using protocol biopsies at day 6.1 ± 2.5. Median follow-up was 7.3 years (IQR 3.9-10.5). Early tacrolimus exposure (<15 days) was evaluated against moderate/severe acute rejection, chronic rejection, graft loss, chronic renal impairment and mortality using multiple logistic and Cox regression. RESULTS: Maintenance immunosuppression was tacrolimus monotherapy (48.1%), double therapy combination with antimetabolites or steroids (18%), or triple therapy combination with antimetabolites and steroids (33.9%). Histological grade of acute rejection was moderate in 157 cases (31.8%) and severe in 19 cases (3.9%). Tacrolimus TC>7 ng/ml on the day of protocol biopsy was associated with less moderate/severe rejection (23.8%) compared with<7 ng/ml (41.2%) (p = 0.004). Mean tacrolimus TC 7-10 ng/ml within 15 days after LT were associated with reduced risk of graft loss (RR = 0.46; p = 0.014) compared to TC 10-15 ng/ml. A peak TC>20 ng/ml within this period was independently related to higher mortality (RR = 1.67; p = 0.005), particularly due to cardiovascular events, infections and malignancy (RR = 2.15; p = 0.001). Early tacrolimus exposure did not influence chronic rejection (p = 0.58), or chronic renal impairment (p = 0.25). CONCLUSIONS: During the first 2 weeks after LT, tacrolimus TC between 7 and 10 ng/ml are safe in terms of acute rejection and are associated with longer graft survival.

Sample size requirement for digital image analysis of collagen proportionate area in cirrhotic livers.

AIMS: The requirements for adequate cirrhotic liver biopsy size have not been established for quantitative fibrosis measurements (collagen proportionate area: CPA). We evaluated the CPA of virtual biopsies in cirrhosis to elucidate (i) the amount of tissue required to achieve reliable CPA measurements and (ii) the effect of aetiology on sample size requirements. METHOD AND RESULTS: A total of 120 cirrhotic tissue blocks (six aetiologies) were studied. A representative 100 mm(2) region was selected from each block and a reference CPA measured. Each image (n = 120) was divided into 100 × 1 mm(2) images; CPA was measured for each 1 mm(2) and virtual biopsies of different sizes were created from the 1 mm(2) components. For each virtual biopsy size the probability that the virtual biopsy CPA would be within 5% of the reference CPA was calculated. There were 441 000 virtual biopsies. Biopsy size versus probability plots indicated that, for 90% probability that the virtual biopsy CPA can be expected to be within 5% of the reference CPA, 22-28 mm(2) of analysable tissue is required depending on liver disease aetiology; and that a 75% probability level requires a biopsy with 12-15 mm(2) of analysable tissue. CONCLUSION: The sample size required for a given probability level depends on the aetiology of cirrhosis, and this should be taken into account when judging the reliability of cirrhotic liver biopsy CPA.

The relationship between transient elastography and histological collagen proportionate area for assessing fibrosis in chronic viral hepatitis.

BACKGROUND: Collagen proportionate area (CPA) has a better correlation with hepatic venous pressure gradient (HVPG) than with Ishak stage. Liver stiffness measurement (LSM) is proposed as non invasive marker of portal hypertension/disease progression. Our aim was to compare LSM and CPA with Ishak staging in chronic viral hepatitis, and HVPG in HCV hepatitis after transplantation. METHODS: One hundred and sixty-nine consecutive patients with chronic hepatitis B virus (HBV) or hepatitis C virus (HCV) infections pre/post liver transplantation (LT), had a liver biopsy combined with LSM (transient elastography), CPA (biopsies stained with Sirius Red and evaluated by digital image analysis and expressed as CPA) and HVPG (measured contemporaneously with transjugular biopsies in LT HCV patients). RESULTS: LSM was dependent on CPA in HBV (r (2) = 0.61, p < 0.0001), HCV (r (2) = 0.59, p < 0.0001) and LT groups (r (2) = 0.64, p < 0.0001). In all three groups, CPA and Ishak were predictors of LSM, but multivariately CPA was better related to LSM (HBV: r (2) = 0.61, p < 0.0001; HCV: r (2) = 0.59, p < 0.0001; post-LT: r (2) = 0.68, p < 0.0001) than Ishak stage. In the LT group, multiple regression analysis including HVPG, LSM, aspartate aminotransferase to platelet ratio index (APRI) and Ishak stage/grade, showed that only CPA was related to HVPG (r (2) = 0.41, p = 0.01), both for HVPG ≥6 mmHg (OR 1.34, 95 % CI 1.14-1.58; p < 0.0001) or ≥10 mmHg (OR 1.25, 95 % CI 1.06-1.47; p = 0.007). CONCLUSION: CPA was related to LSM in HBV or HCV hepatitis pre/post-LT. CPA was better related to LSM than Ishak stage. In the LT HCV group, CPA was better related to HVPG than Ishak stage/grade, LSM or APRI. CPA may represent a better comparative histological index for LSM, rather than histological stages.

A systematic review of microvascular invasion in hepatocellular carcinoma: diagnostic and prognostic variability.

Selected patients with hepatocellular carcinoma are candidates to receive potentially curative treatments, such as hepatic resection or liver transplantation, but nevertheless there is a high risk of tumor recurrence. Microvascular invasion is a histological feature of hepatocellular carcinoma related to aggressive biological behavior. We systematically reviewed 20 observational studies that addressed the prognostic impact of microvascular invasion, either after liver transplantation or resection. Outcomes were disease-free survival and overall survival. In liver transplantation, the presence of microvascular invasion shortened disease-free survival at 3 years (relative risk (RR)=3.41 [2.05-5.7]; five studies, n=651) and overall survival both at 3 years (RR=2.41 [1.72-3.37]; five studies, n=1,938) and 5 years (RR=2.29 [1.85-2.83]; six studies, n=2,003). After liver resection, microvascular invasion impacted disease-free survival at 3 and 5 years (RR=1.82 [1.61-2.07] and RR=1.51 [1.29-1.77]; four studies, n=1,501 for both comparisons). However inter/intraobserver variability in reporting and the lack of definition and grading of microvascular invasion has led to great heterogeneity in evaluating this histological feature in hepatocellular carcinoma. Thus, there is an urgent need to clarify this issue, because determining prognosis and response to therapy have become important in the current management of hepatocellular carcinoma. In this systematic review, we summarize the diagnostic and prognostic data concerning microvascular invasion in hepatocellular carcinoma and present a basis for consensus on its definition.

Liver collagen proportionate area predicts decompensation in patients with recurrent hepatitis C virus cirrhosis after liver transplantation.

BACKGROUND AND AIMS: Current histological scoring systems do not subclassify cirrhosis. Computer-assisted digital image analysis (DIA) of Sirius Red-stained sections measures fibrosis morphologically producing a fibrosis ratio (collagen proportionate area [CPA]). CPA could have prognostic value within a disease stage, such as cirrhosis. The aim of the present study was to evaluate CPA in patients with recurrent hepatitis C virus (HCV) allograft cirrhosis and assess its relationship with hepatic venous pressure gradient (HVPG). METHODS: In 121 consecutively-transplanted HCV patients with HVPG, measured contemporaneously with transjugular liver biopsies, 65 had Ishak stage 5 or 6 disease (43 with HVPG measurement). Biopsies were stained with Sirius Red for DIA, and the collagen content was expressed as a CPA. In three cases, a tissue for Sirius Red staining was not obtained, and the patients were excluded. RESULTS: Sixty-two patients were analyzed. The median HVPG was 8 mmHg (interquartile range [IQR]: 5-10). Portal hypertension (HVPG ≥ 6 < 10 mmHg) was present in 30 (69.8%), and HVPG ≥ 10 mmHg in 13 (30.2%). The median CPA was 16% (IQR 10.75-23.25). Median Child-Pugh score and HVPG were not significantly different between Ishak fibrosis stage 5 or 6, whereas CPA was statistically different: 13% in stage 5 (IQR 8.3-12.4) versus 23% in stage 6 (IQR 17-33.7, P < 0.001). In the multivariate analysis, CPA was the only variable significantly associated with clinically-significant portal hypertension (HVPG ≥ 10 mmHg, odds ratio: 1.085, confidence interval: 1.004-1.172, P = 0.040). A CPA of 14% was the best cut-off value for clinically-significant portal hypertension (CSPH) and liver decompensation, which occurred in 24 patients. Event-free survival was significantly shorter in patients with CSPH or with a CPA value ≥ 14%, or with a combination of both. CONCLUSION: In Ishak stages 5 and 6, CPA correlated with HVPG, but had a wider range of values, suggesting a greater sensitivity for distinguishing "early" from "late" severe fibrosis/cirrhosis. CPA was a unique, independent predictor of HVPG ≥ 10 mmHg. CPA can be used to subclassify cirrhosis and for prognostic stratification.

Predicting severity and clinical course of acute rejection after liver transplantation using blood eosinophil count.

Acute cellular rejection remains an important source of morbidity after liver transplantation, particularly if rejection is moderate or severe, as this usually is treated. Currently liver biopsies are seldom performed, so diagnostic noninvasive markers would be useful. We evaluated 690 consecutive first liver transplant patients to assess whether peripheral eosinophilia could predict moderate-severe rejection and its course. A protocol biopsy was performed 6 ± 2.5 days after transplant. A second biopsy was taken 6.1 ± 2 days after the first in 487 patients to assess histological improvement. Liver function tests, peripheral eosinophil count and changes between first and second biopsy, were evaluated using logistic regression. Histological rejection was present in 532 patients (77.1%), with moderate (30.6%) and severe rejection (3.9%). Peripheral eosinophil count was strongly associated with moderate-severe rejection (OR = 2.15; P = 0.007), although the area under ROC curve (AUROC) was 0.58. On second biopsy, rejection improved in 119 (24.4%) patients. The delta in eosinophil count between the first and second biopsies was the only independent predictor of histological improvement (OR = 3.12; P = 0.001), irrespective of whether bolus steroids were used (OR = 2.77; P = 0.004); AUROC was 0.72. Peripheral eosinophilia is not sufficiently predictive of moderate-severe histological rejection. However the changes in eosinophil count over time can accurately predict the histological resolution of rejection.

Renal failure and cirrhosis: a systematic review of mortality and prognosis.

BACKGROUND & AIMS: To evaluate renal failure (RF) in cirrhosis to determine and quantify its prognostic significance. METHODS: Studies were identified by MEDLINE, EMBASE, Cochrane, ISI Web of Science (1977-2010); search terms included renal failure, mortality, and cirrhosis. Included studies (n=74) reported >10 patients and mortality data (8088 patients). Mortality at 1, 3, and 12 months was evaluated with respect to Child-Pugh score, serum creatinine, ascites, ICU status or sepsis, prospective study design, and publication year. Pooled odds ratio (POR) for death was compared for RF vs. non-RF (5668 patients). RESULTS: Overall median mortality for RF patients was 67%: 58% at 1 month and 63% (IQR 54-79) at 12 months. POR for death RF vs. non-RF patients was 7.6 (95%CI 5.4-10.8). Overall mortality before 2005 (1264 patients) was 74% and after 2005 (2833 patients) was 63% with a marked reduction only at 30 days (71% vs. 52%). CONCLUSIONS: This study provides a measure of the increased risk of death in cirrhosis with renal failure. RF increases mortality 7-fold, with 50% of patients dying within one month. Preventative strategies for RF are needed.

Pitfalls in assessing renal function in patients with cirrhosis--potential inequity for access to treatment of hepatorenal failure and liver transplantation.

Serum creatinine is universally used to assess renal function in clinical practice. Creatinine and changes in serum creatinine are used to define acute kidney injury and hepatorenal syndrome (HRS) in patients with progressive liver disease. In addition, creatinine is a key variable in the calculation used to determine priority for liver transplantation in many countries. As there is no universal standardized creatinine assay, there is variation in creatinine determinations between laboratory assays, compounded by assay interference due to chromogens, including bilirubin. This leads to patients with the same actual renal function potentially being offered different treatment options, in terms of access to therapy for HRS and priority waiting time for liver transplantation. Alternative methods for assessing renal function either also tend to overestimate renal function or are too time consuming and expensive to provide practical alternatives for standard clinical practice. Standardization of creatinine assays with readily available reference standards would help minimize interlaboratory variation; of the current creatinine assays, enzymatic creatinine appears more accurate, but even this is inaccurate at high bilirubin concentrations. Further work is required to determine whether interpatient variation can be reduced by correcting creatinine and cystatin measurements for muscle mass.

Comparison of cystatin C and creatinine-based glomerular filtration rate formulas with 51Cr-EDTA clearance in patients with cirrhosis.

BACKGROUND AND OBJECTIVES: Renal function is an important predictor of survival in cirrhosis and liver transplantation. GFR estimates using serum cystatin C (CysC) are proposed as better predictors of renal function than ones on the basis of serum creatinine (Cr). Our aims were: (1) evaluate correlations between serum CysC and different methods of creatinine measurements; (2) compare CysC and Cr GFR formulas with (51)Cr-EDTA; and (3) evaluate liver-related parameters potentially influencing GFR. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: 254 blood samples in 65 patients with cirrhosis correlating CysC with four Cr methods were used; another 74 patients comparing (51)Cr-EDTA GFR to Modification of Diet in Renal Disease and Larsson and Hoek formulas for CysC were also included. Agreement was assessed using Bland-Altman plots and concordance correlation coefficients. Multivariate linear regression analysis was used for GFR predictors. RESULTS: Serum CysC correlated modestly with O'Leary modified Jaffe, compensated kinetic Jaffe, enzymatic creatinine, and standard kinetic Jaffe 0.72/0.71/0.72/0.72 (all P < 0.001). Bland-Altman agreement with (51)Cr-EDTA GFR was poor; the best agreement was Modification of Diet in Renal Disease (concordance 0.61; 95% CI, 0.47 to 0.71); the worst agreement was the Hoek formula (concordance 0.46; 95% CI, 0.27 to 0.61). A new GFR formula including the Child-Pugh score improved the accuracy of Cr GFR formulas compared with (51)Cr-EDTA GFR. CONCLUSIONS: Estimated GFR in cirrhosis is not better with CysC formulas compared with creatinine ones: specific formulas may be necessary.

Assessment of adrenocortical reserve in stable patients with cirrhosis.

BACKGROUND & AIMS: Adrenal insufficiency (AI) is reported in critically ill patients with cirrhosis and is associated with increased mortality. It is unclear if AI is an underlying condition or triggered by critical events (e.g. sepsis). We investigated AI in cirrhosis without infection or hemodynamic instability. METHODS: A total of 101 consecutive patients with cirrhosis were studied. AI was defined by a total serum cortisol (TC) <18 µg/dl at 20 or 30 min after injection of 1 µg of tetracosactrin. Transcortin, calculated free cortisol (cFC), and free cortisol index (FCI) were assessed in a subgroup of 41 patients, with FCI>12 representing normal adrenal function. RESULTS: AI was present in 38 patients (38%). Child score (median, 10 vs 7, p<0.0001), MELD score (median, 17 vs 12, p<0.0001), ascites (68% vs 37%, p<0.01), basal TC (median,7.6 vs 14.9 µg/dl, p<0.001), albumin (28 ± 0.8 vs 33 ± 0.7 g/L, p<0.0001), INR (median, 1.6 vs 1.2, p<0.0001), total bilirubin (median, 51 vs 31 µmol/L, p<0.05), total cholesterol (median, 120 vs 142, p<0.05), and LDL (median, 76 vs 81, p<0.05) were significantly different between those with and without AI. ROC curves showed a basal TC ≤ 12.8 µg/dl to be a cut-off value closely associated with AI. The cFC was significantly related to TC for baseline values (R=0.94, p<0.0001), peak values (R=0.90, p<0.0001), and delta values (R=0.95, p<0.0001), in patients with and without AI. However, no patient had a FCI<12. CONCLUSIONS: AI defined by an abnormal response to 1 µg tetracosactrin is frequent in stable patients with cirrhosis, in the absence of infections or hemodynamic instability and is related to the severity of liver disease. However, evaluation of the true incidence of AI should comprise direct assays of free cortisol. Clinical consequences of AI need to be explored.