High Preoperative Recipient Plasma 7␤-hydroxycholesterol Is Associated With Initial Poor Graft Function After Liver Transplantation Stefano Ginanni Corradini,1 Fausta Micheletta,2 Silvia Natoli,2 Massimo Iappelli,3 Emanuele Di Angelantonio,2 Rosanna De Marco,1 Walter Elisei,1 Maria Siciliano,1 Massimo Rossi,3 Pasquale Berloco,3 Adolfo Francesco Attili,1 Ulf Diczfalusy,4 and Luigi Iuliano5 Oxidative stress is implicated in the pathogenesis of hepatic ischemia-reperfusion injury, a major determinant of initial poor graft function (IPGF) after orthotopic liver transplantation (OLT). We prospectively investigated the association between the recipient plasma preoperative oxidative stress and the occurrence of IPGF after deceased-donor OLT and indirectly studied the source— hepatic or extra-hepatic— of systemic oxidative stress in vivo in cirrhosis. We used a recently developed specific and sensitive mass spectrometry assay to measure 7␤-hydroxycholesterol and 7-ketocholesterol (oxysterols), markers of oxidative stress, in biological matrices. At univariate analysis, preoperative recipient 7␤-hydroxycholesterol plasma concentration was significantly higher in transplants with subsequent IPGF (n ⴝ 9) compared with those with initial good graft function (IGGF; n ⴝ 23) [mean ⴞ SD: 30.63 ⴞ 26.42 and 11.57 ⴞ 15.76 ng/mL, respectively] (P ⴝ 0.017). In a logistic regression model, which included also the Model for End-Stage Liver Disease (MELD) score, 7␤-hydroxycholesterol plasma concentration was an independent predictor of IPGF with an odds ratio of 1.17 (95% CI, 1.02-1.33, P ⴝ 0.028). Patients with cirrhosis (n ⴝ 32) had increased oxysterol plasma levels compared with healthy controls (n ⴝ 49); livers with cirrhosis (n ⴝ 21), however, had oxysterol content comparable with normal livers obtained from organ donors (n ⴝ 19). Oxysterols persisted elevated in plasma 1 month after OLT (n ⴝ 23). In conclusion, cir- Abbreviations: IPGF, initial poor graft function; IGGF, initial good graft function; MELD, Model for End-Stage Liver Disease; OLT, orthotopic liver transplantation; ROS, reactive oxygen species; ALT, alanine aminotransferase; HCV, hepatitis C virus. From the Departments of 1Clinical Medicine, Division of Gastroenterology, 2Internal Medicine, and 3Surgical Sciences, University La Sapienza, Rome, Italy; the 4Department of Laboratory Medicine, Karolinska University Hospital, Huddinge, Sweden; and the 5Department of Internal Medicine, University La Sapienza at Polo Pontino-Latina, Italy. Received March 8, 2005; accepted May 27, 2005. S.G.C. and F.M contributed equally to the manuscript. Address reprint requests to Luigi Iuliano, MD, Department of Internal Medicine, University La Sapienza, Via del Policlinico 155, 00161 Rome, Italy. Telephone/FAX: (39) 06 4997 7777; E-mail: luigi.iuliano@uniroma1.it Copyright © 2005 by the American Association for the Study of Liver Diseases Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/lt.20524 1494 rhosis presents upregulated systemic oxidative stress likely of extrahepatic source that is associated with graft failure after OLT. (Liver Transpl 2005;11:1494-1504.) T he common use of marginal grafts for OLT is associated with an increased risk of initial poor graft function (IPGF) and reduced survival.1,2 Several studies have shown that some donor, graft, perioperative, and operative variables may influence graft dysfunction; therefore, to reduce the risk of IPGF, the search for new variables associated with graft dysfunction after human orthotopic liver transplantation (OLT) is continually needed.1,2 Recently, the severity of recipient liver disease as expressed by the Model for End-Stage Liver Disease (MELD) score also has been shown to be associated with 3-month OLT survival.3 Oxidative stress at reperfusion is thought to play a major pathogenetic role in hepatic ischemia-reperfusion injury, the underpinning of early graft dysfunction1,4; in fact, in experimental models reactive oxygen species (ROS) released at reperfusion by activated Kupffer cells recruited neutrophils and hepatocytes induced a series of events at a cellular level, culminating with apoptosis and necrosis and production of inflammatory mediators that provoke circulatory disturbances and worsening of liver damage.1,4-6 We hypothesized that unlike experimental OLT where the recipient is healthy, in human OLT the biochemical machinery that determines the severity of ischemia-reperfusion injury could also be influenced by the pre-existing altered blood oxidative stress/antioxidant status of the recipient, which eventually persists after removal of the liver with cirrhosis. Our hypothesis is supported by two considerations. First, oxidative stress is involved in the pathogenesis of chronic liver disease and fibrosis, and increased oxidative stress with impaired antioxidant status at the systemic level has been described in different chronic liver diseases.7-23 Second, two previous studies have shown the persistence of increased systemic lipid peroxidation products above normal values in patients with cirrhosis 3 to 12 weeks after OLT that is a likely Liver Transplantation, Vol 11, No 12 (December), 2005: pp 1494-1504 1495 7␤-hydroxycholesterol and IPGF expression of extrahepatic ROS production. In fact, this time frame from operation is too wide to be compatible with continuous intra-graft production of ROS secondary to ischemia-reperfusion injury and release into the circulation, and it is too small to be expression of chronic liver damage of the transplanted liver.24,25 The main aim of the present study was to investigate the relationships between the preoperative plasma oxidative stress/antioxidant status of patients with cirrhosis and the occurrence of IPGF after “non-status 1,” fullsize liver, primary, deceased donor OLT. We choose to measure 7␤-hydroxycholesterol and 7-ketocholesterol as markers of oxidative stress because they are free radical – mediated cholesterol oxidation products (oxysterols) identified as valuable in vivo systemic oxidative stress markers.26 In addition, we recently developed a method for the simultaneous evaluation of these oxysterols and ␣-tocopherol in the same sample as a measure of systemic oxidative stress/antioxidant status.27 Finally, we choose to measure systemic oxidative stress by plasma 7␤-hydroxycholesterol and 7-ketocholesterol because, in analogy to their injurious effect on arterial endothelium in atherosclerosis, their concentration in plasma perfusing the graft after OLT could directly influence sinusoidal endothelial cell damage, a key step of ischemia-reperfusion injury.1,4,28 The second aim of the present study was to evaluate the source— hepatic or extra-hepatic— of systemic oxidative stress in vivo in cirrhosis. This was pursued by measuring the oxidative stress/antioxidant status in plasma in healthy controls, in patients with cirrhosis before and 1 month after OLT, and at the hepatic tissue level in cirrhosis and normal livers. The final aim was to verify whether the systemic oxidative stress/antioxidant status imbalance correlates with the severity of chronic liver disease. Patients and Methods Study Population and Design Between February 2001 and February 2003, 59 consecutive non-status 1, full-size liver, primary, deceased donor transplantations were performed at the Liver Transplantation Unit of University “La Sapienza” Hospital in Rome. Twenty-seven patients were excluded from the study because a blood sample taken within 4 weeks before surgery was not available. This exclusion criteria was adopted in order to investigate the relation between the preoperative plasma oxidative stress/antioxidant status of patients with cirrhosis and the occurrence of IPGF after OLT. Oxysterol and ␣-tocopherol measurement needed to be performed as close as possible to the transplantation day on blood samples obtained between 8:00 and 9:00 Table 1. Parameters Used for the Calculation of Scoring System Assessing Early Postoperative Graft Function in Liver Transplant Patients Parameter Serum ALT (U/L)* ⬍1,000 1,000 – 2,500 ⬎2,500 Bile output (mL/day)† ⬎100 40 – 100 ⬍40 Prothrombin activity (%)‡ ⬎60 (spontaneous) ⬎60 (with fresh-frozen plasma) ⬍60 (with fresh-frozen plasma) Assigned Value 1 2 3 1 2 3 1 2 3 NOTE. IPGF ⫽ score 7 – 9. IGGF ⫽ score 3 – 6. * Highest value in the 72 hours after liver transplantation. † Mean value during the first 72 hours after liver transplantation. ‡ Lowest value in the first 72 hours after liver transplantation. A.M. after an overnight fasting. Donor livers with macroscopic fatty infiltration confirmed by a biopsy with more than 50% of macrovesicular steatosis were not utilized. Donor arterial blood gas analysis was obtained between 4 hours and 1 hour before organ harvesting. The immunosuppressive protocol was based on a triple therapy with cyclosporine microemulsion, methylprednisolone, and mycophenolate mofetil. Blood concentration of cyclosporine microemulsion was monitored at time 0 and at 2 hours after ingestion. Methylprednisolone was rapidly tapered. To assess the quality of early postoperative graft function, we classified the patients with cirrhosis who underwent transplantation according to the score developed by Gonzalez et al.29 This score was obtained from the peak alanine transaminase (ALT) value, mean bile output, and lowest prothrombin activity measured in the first 72 postoperative hours (Table 1). In each patient the score was calculated from the sum of the assigned values for each parameter. On the basis of this score, patients with a score ranging from 7 to 9 were considered with IPGF, while patients with a score ranging from 3 to 6 were considered with initial good graft function (IGGF).2,29 Primary nonfunction was defined as a nonrecoverable hepatocellular function necessitating emergency retransplantation within 7 days. To investigate the effects of the healthy liver implanted at OLT on systemic oxidative stress and antioxidant status, a further blood sample was taken in a subgroup of the OLT patients after a follow-up of 25 to 35 days after transplantation to avoid potential artifacts associated with surgery. This was done in 23 patients only, since 1 patient died 6 days after surgery; 2 patients were excluded because they underwent retransplantation 16 and 17 days after initial transplantation, respectively; and 6 patients did not return for posttransplan- 1496 Corradini et al. tation blood withdrawal in the scheduled time frame between days 25 and 35. To investigate whether systemic oxidative stress/antioxidant status imbalance correlates with the severity of chronic liver disease, pretransplantation plasma oxidative stress/antioxidant status of the patients with cirrhosis (n ⫽ 32) was compared to that of patients with hepatitis C virus (HCV) chronic hepatitis (n ⫽ 19) and age- and sex-matched healthy controls (n ⫽ 49). Diagnoses of chronic hepatitis and liver cirrhosis were established on the basis of histology. Healthy subjects were recruited from the hospital staff and their relatives who did not show any clinical or biochemical signs of disease. Exclusion criteria for all patients of this study were prior stomach surgery, cancer (other than hepatocellular carcinoma), and malabsorption syndrome. Neither patients nor controls used any drugs or supplements containing vitamin E, vitamin C, carotenoids, or iron for at least 30 days before the enrollment and during the study period. Venous blood samples were taken after fasting at least 12 hours between 8:00 and 9:00 A.M. for the evaluation of routine biochemical analysis, oxysterols, and ␣-tocopherol. In order to compare the degree of tissue oxidative stress and antioxidant status in the normal liver and liver with cirrhosis, we measured oxysterols and ␣-tocopherol in tissue samples from 21 of the explanted livers with cirrhosis at OLT and from 19 heartbeating organ donors from whom a preexplant biopsy was obtained before the aorta was clamped. The study protocol was approved by the local ethics committee, formally Comitato Etico Azienda Policlinico Umberto I, Rome, Italy. All subjects gave informed consent to participate in the study. There is no chance that a patient (whether living, dead, or a child) may be identified from the present paper. Blood Analysis Plasma was stored at ⫺80° for ␣-tocopherol, 7␤-hydroxycholesterol, and 7-ketocholesterol measurements. Plasma levels of ␣-tocopherol were analyzed by HPLC, and oxysterols were measured from the same sample by mass spectrometry using an isotope dilution method as previously described.27 Serum cholesterol, aspartate transaminase (AST), ALT, alkaline phosphatase (ALP), ␥-glutamyl-transpeptidase (␥GT), albumin, bilirubin, prothrombin activity (INR), and creatinine were determined by standard automated techniques. Tissue Analysis To measure oxysterols and ␣-tocopherol in hepatic tissue, immediately after collection liver biopsies were washed of contaminating blood with saline, put on filter paper to absorb liquid, and then weighed and frozen in liquid nitrogen until assay. On the day of assay, tissue samples were finely minced and transferred with 10 mL chloroform/methanol (2:1 vol/ vol) solution containing 0.01% butylated hydroxytoluene (vol/wt) to a glass homogenizer. ␣-Tocopherol acetate and deuterium-labeled cholesterol; 7␤-hydroxycholesterol and 7-ketocholesterol as internal standards; and butylated hydroxytoluene and ethylene diamine tetra acetate as an antioxidant and a metal-complexing reagent, respectively, were added. ␣-Tocopherol, cholesterol, and oxysterols were measured in the same sample by a recently developed method based on gas chromatography/mass spectrometry.27 The mass spectrometer was operated in the selected ion monitoring mode, and the ions used for analysis were as follows: [2H6]cholesterol, 335 m/z; cholesterol, 329 m/z; [2H6]7␤hydroxycholesterol, 462 m/z; 7␤-hydroxycholesterol, 456 m/z; [2H6]7-ketocholesterol, 478 m/z; 7-ketocholesterol, 472 m/z; ␣-tocopherol, 502 m/z; ␣-tocopherol acetate, 430 m/z. Protein concentration in liver homogenates was measured by the Bio-Rad Protein Assay method (BIO-RAD, Munchen, Germany). Aliquots of pre-explant allograft biopsy obtained from heartbeating donors before the aorta was clamped were immediately fixed in formalin for steatosis assessment. The steatosis level was blindly scored independently by two of the authors (A.F.A. and S.N.). They measured the percentage of hepatocytes with microvesicular and macrovesicular steatosis. Data were then analyzed according to the Australian National Liver Transplantation Unit scale of macrovesicular steatosis, which is as follows: S0 ⫽ steatosis absent, S1 ⫽ less than 30%, S2 ⫽ 30% to 60%, and S3 ⫽ more than 60% of hepatocytes with steatosis.30,31 Statistical Analysis Analysis of data was carried out using the SPSS software (SPSS version 11.0, Chicago, IL). For continuous variables such as age and laboratory parameters, descriptive statistics were calculated and reported as mean ⫾ standard deviation. Normalcy of distribution of continuous variables was assessed using the Kolmogorov-Smirnov test. Normally distributed continuous variables were compared by group using one-way analysis of variance (ANOVA) followed post hoc by Bonferroni’s test. Continuous variables with distributions deviating significantly from normal were compared by group using the Kruskal-Wallis test followed post hoc by the Mann-Whitney U test. Liver biopsy data were compared between recipients and donors using the t test for independent samples. Pre- and posttransplantation data in recipients were compared using the paired t test. Categorical variables such as sex and the presence of specific medical conditions were described using frequency distributions. The chi-square test with 99% Monte Carlo confidence intervals was used to detect differences in categorical variables by group. Pretransplantation indicators of oxidative stress were compared by subsequent initial graft function using the t test for independent samples. Initial graft function was modeled using stepwise logistic regression analysis, and odds ratios with 95% confidence limits were calculated. Spearman correlation test was used because non-normally distributed variables were analyzed. All tests were twosided and considered significant at P ⬍ 0.05. 7␤-hydroxycholesterol and IPGF Figure 1. (A) Positive correlation between recipient preoperative 7␤-hydroxycholesterol plasma concentration and postoperative 3-day peak serum ALT after OLT (r ⴝ 0.355, P < 0.05), and (B) negative correlation between recipient preoperative 7␤-hydroxycholesterol plasma concentration and the mean 3-day postoperative bile output (r ⴝ ⴚ0.497, P < 0.01). Results Recipient Pretransplantation Plasma Oxysterol and Vitamin E Concentrations in Relation to the Quality of Initial Graft Function According to the Gonzalez criteria for the assessment of early postoperative graft function, of the patients with cirrhosis who were submitted to OLT 9 were categorized as IPGF and 23 as IGGF. At postoperative day 1, the IPGF as compared to the IGGF group had significantly (P ⬍ 0.001) higher serum ALT (1182.1 ⫾ 635.1 and 491.5 ⫾ 373.3 U/L, respectively) and AST (1678.6 ⫾ 1463.1 and 661.2 ⫾ 412.8 U/L, respectively) concentrations. The same intergroup differences were found at postoperative days 2 and 3 (data not shown). The mean bile output during the first 72 postoperative hours was significantly lower in the IPGF group than in the IGGF group (54.4 ⫾ 35.7 and 137.8 ⫾ 74.6 mL/d, respectively; P ⬍ 0.01). Preoperative recipient 7␤-hydroxycholesterol plasma concentration correlated positively (r ⫽ 0.355, P ⬍ 0.05) with postoperative 3-day peak serum ALT (Fig. 1A) and negatively (r ⫽ ⫺0.497, P ⬍ 0.01) with the mean 3-day postoperative bile output (Fig. 1B). Preoperative 1497 7-ketocholesterol plasma concentration inversely correlated (r ⫽ ⫺0.398, P ⬍ 0.05) with the mean 3-day postoperative bile output. Table 2 shows the results of the univariate analysis of the different variables analyzed in relation to the quality of initial graft function. The IPGF group showed significantly higher recipient preoperative plasma concentrations of 7␤-hydroxycholesterol compared with the IGGF group. The 2 patients who required retransplantation— one for a primary nonfunction and the other for hepatic artery thrombosis—showed the highest 7␤-hydroxycholesterol plasma concentrations of the entire series before their primary transplantations (81.3 and 79.7 ng/mL, respectively). The IPGF group showed higher recipient preoperative plasma concentrations of 7-ketocholesterol compared with the IGGF group, at the limit of statistical significance. No significant difference in recipient preoperative ␣-tocopherol plasma concentration was observed between the IPGF and the IGGF groups. Although the preoperative oxysterol to ␣-tocopherol ratio in plasma was 71.5% higher in patients with IPGF than in patients with IGGF, the difference was not statistically significant. The IPGF group showed significantly lower donor preharvest arterial partial pressure of oxygen (PaO2) compared with the IGGF group. Although the recipient MELD score tended to be higher and the use of noradrenaline in donors was more frequent in the IPGF than in the IGGF group, these differences were not statistically significant. The IPGF and the IGGF groups did not differ in terms of donor age, cardiac arrest, cause of brain death, serum sodium concentration, length of intensive care stay, and graft cold ischemia time. In a logistic regression model, the only variable independently associated with IPGF was 7␤-hydroxycholesterol with an odds ratio of 1.17 (95% CI, 1.02-1.33; P ⫽ 0.028); for every 1-ng/mL increase in plasma 7␤-hydroxycholesterol, odds of poor graft function increased by 17%. Graft steatosis score was available in 7 grafts of the IPGF group and in 15 grafts of the IGGF group only, and did not differ in the two groups (Table 2). Changes of Plasma Oxysterol and Vitamin E Concentrations After OLT When recipient plasma concentration of 7␤-hydroxycholesterol and 7-ketocholesterol before and at 1 month after OLT were compared, a postoperative nonsignificant increase in plasma 7␤-hydroxycholesterol concentration (13.68 ⫾ 14.27 to 17.43 ⫾ 20.8 ng/mL, before and after OLT, respectively), and 7-ketocholesterol (17.71 ⫾ 12.40 to 28.96 ⫾ 21.36 ng/mL, before 1498 Corradini et al. Table 2. Univariate Analysis of Recipient and Donor Variables in OLTs with IPGF or IGGF Recipient preoperative 7␤-hydroxycholesterol (ng/mL)* 7-ketocholesterol (ng/mL)* ␣-tocopherol (mg/dL)* oxysterols/␣-tocopherol*† MELD score Donor Age (yrs) Cardiac arrest, number (%) Yes No Cause of brain death, number (%) Traumatic Vascular/anoxic Infusion with noradrenaline, number (%) Yes No Serum sodium (mEq/L) Intensive care unit stay (days) Preharvest PaO2 (mmHg) Graft cold ischemia time (min) Graft macrovesicular steatosis‡ S0 S1 S2 S3 IPGF IGGF P 30.63 ⫾ 26.42 24.21 ⫾ 12.30 0.87 ⫾ 0.32 68.8 ⫾ 39.3 17.0 ⫾ 10.7 11.57 ⫾ 15.76 14.24 ⫾ 11.12 0.71 ⫾ 0.29 40.1 ⫾ 29.3 13.2 ⫾ 5.6 0.017 0.060 0.198 0.091 0.146 49.2 ⫾ 18.1 47.0 ⫾ 17.9 0.751 0.361 0 (0) 7 (100) 2 (9) 21 (91) 2 (22) 7 (78) 5 (22) 18 (78) 3 (33) 6 (67) 150.3 ⫾ 14.7 4.9 ⫾ 4.7 107.4 ⫾ 48.0 397.0 ⫾ 83.0 4 (17) 19 (83) 154.0 ⫾ 13.1 3.7 ⫾ 2.8 161.8 ⫾ 57.1 380.7 ⫾ 79.1 0.976 0.327 4 (57) 2 (29) 1 (14) 0 (0) 0.491 0.362 0.018 0.608 0.841 9 (60) 5 (33) 1 (7) 0 (0) * Plasma value. † (7␤-hydroxycholesterol ⫹ 7-ketocholesterol)/␣-tocopherol. ‡ Available in 7 grafts of the IPGF and in 15 grafts of the IGGF group only. and after OLT, respectively) was found. As shown in Fig. 2, the increase in plasma oxysterols after OLT was largely dependent on the contribution of HCV carriers (Fig. 2A,C), rather than of HCV-negative patients (Fig. 2B,D). The mean 7␤-hydroxycholesterol concentration after OLT increased by 54% in HCV-positive (11.95 vs. 18.39 ng/mL, before and after OLT, respectively),compared with a slight reduction (⫺4%) in HCV-negative patients (16.37 vs. 15.75 ng/mL, before and after OLT, respectively). The mean 7-ketocholesterol increased by 114% in HCV-positive (14.79 vs. 31.64 ng/mL, before and after OLT, respectively), compared with 11% in HCV-negative patients (22.46 vs. 24.94 ng/mL, before and after OLT, respectively). These differences, however, were not statistically significant. When recipient plasma concentration of ␣-tocopherol before and 1 month after OLT were compared, a postoperative significant (P ⬍ 0.0002) increase (0.77 ⫾ 0.31 to 1.16 ⫾ 0.50 mg/dL, mean ⫾ SD, before and after OLT, respectively) was found. As shown in Fig. 3, plasma ␣-tocopherol significantly increased after OLT in both HCV carriers (0.71 ⫾ 0.38 to 1.20 ⫾ 0.59 mg/dL, before and after OLT, respectively; Fig. 3A) and in HCV-negative patients (0.87 ⫾ 0.15 to 1.10 ⫾ 0.35 mg/dL, before and after OLT, respectively; Fig. 3B). The oxysterol/␣-tocopherol ratio in plasma did not change after OLT when the entire study population (42.5 ⫾ 29.9 and 38.4 ⫾ 31.6, before and after OLT, respectively), or HCV carriers (43.6 ⫾ 27.8 and 40.8 ⫾ 37.7, before and after OLT, respectively), or HCVnegative patients (40.7 ⫾ 34.6 and 34.6 ⫾ 20.2, before and after OLT, respectively), were considered. Plasma Oxysterol and Vitamin E Concentrations in Relation to Staging of Chronic Liver Disease The baseline characteristics of the patients with cirrhosis who were submitted to OLT and of the chronic hepatitis and control groups are shown in Table 3. 7␤-hydroxycholesterol and IPGF 1499 Figure 2. Box plots of plasma concentrations of oxysterols before and 1 month after OLT according to HCV status; 7␤-hydroxycholesterol in HCV-positive (A) and HCV-negative (B) patients; 7-ketocholesterol in HCV-positive (C) and HCV-negative (D) patients. Patients with cirrhosis were significantly younger than patients with chronic hepatitis. There were no differences in sex and smoking habits by group. Mean plasma cholesterol concentration was significantly lower in patients with cirrhosis than in the other two groups. Table 4 shows the oxysterol and ␣-tocopherol plasma concentrations in the control and chronic hepatitis subjects and in the patients with cirrhosis before OLT. Patients with chronic hepatitis exhibited signifi- cantly higher 7␤-hydroxycholesterol and 7-ketocholesterol plasma levels than controls. Plasma levels of oxysterols were significantly higher in patients with liver cirrhosis compared with either controls or patients with chronic hepatitis. Plasma ␣-tocopherol concentrations were significantly lower in patients with liver cirrhosis than controls. Although plasma ␣-tocopherol levels were 19.3% lower in patients with cirrhosis than in subjects with chronic hepatitis, this difference did not reach statistical Figure 3. Box plots of plasma concentrations of ␣-tocopherol in HCV-positive (A) and HCV-negative (B) patients with cirrhosis before and 1 month after OLT. *P < 0.05; **P < 0.005. 1500 Corradini et al. Table 3. Baseline Characteristics of the Patients With Cirrhosis Submitted to OLT and of the Chronic Hepatitis and Control Groups Age, yrs Male sex, n (%) Smoking habit, n (%) Prothombin activity (INR) Albumin, g/dL Total bilirubin, mg/dL AST, U/L ALT, U/L Cholesterol, mg/dL Creatinine, mg/dL Etiology of cirrhosis, n (%) HCV HBV Ethanol Mixed¶ PBC Criptogenetic Presence of HCC, n (%) Controls n ⫽ 49 Chronic Hepatitis n ⫽ 19 Liver Cirrhosis n ⫽ 32 56.3 ⫾ 13.7 28 (57) 14 (29) 0.98 ⫾ 0.10 4.2 ⫾ 0.6 0.80 ⫾ 0.20 24.5 ⫾ 8.2 19.5 ⫾ 9.1 213.6 ⫾ 37.5 0.90 ⫾ 0.15 61.1 ⫾ 13.7 10 (53) 6 (32) 1.05 ⫾ 0.07 4.1 ⫾ 0.4 0.83 ⫾ 0.46 45.6 ⫾ 15.7§ 58.8 ⫾ 48.0§ 212.8 ⫾ 38.2 0.89 ⫾ 0.21 54.4 ⫾ 7.1* 22 (69) 10 (32) 1.63 ⫾ 0.58‡§ 3.2 ⫾ 0.4‡§ 5.26 ⫾ 8.21‡§ 69.6 ⫾ 42.8†§ 52.5 ⫾ 30.9§ 144.7 ⫾ 47.5‡§ 0.90 ⫾ 0.31 — — — — — — — 19 (100) — — — — — — 17 (53) 6 (19) 2 (6) 3 (9) 1 (3) 3 (9) 10 (31) * P ⬍ 0.05 vs. chronic hepatitis. † P ⬍ 0.01 vs. chronic hepatitis. ‡ P ⬍ 0.001 vs. chronic hepatitis. § P ⬍ 0.001 vs. controls. ¶ Ethanol plus HCV or HBV. significance. Furthermore, plasma ␣-tocopherol concentration was significantly lower in chronic hepatitis patients than controls. The oxysterol to ␣-tocopherol ratio in plasma was significantly higher in patients with cirrhosis compared with either controls or patients with chronic hepatitis. Additionally, patients with chronic hepatitis showed significantly higher levels of the oxy- sterols to ␣-tocopherol ratio in plasma compared with controls. No difference was observed in terms of plasma oxysterols and ␣-tocopherol in the patients with cirrhosis when they were subgrouped according to either disease severity (MELD ⱖ 20 vs. MELD ⬍ 20; Table 4) or their HCV status (HCV-positive vs. HCV-negative; data not shown). Table 4. Oxysterol and ␣-tocopherol Plasma Concentrations and Oxysterol to ␣-tocopherol Ratio in Plasma of the Cirrhosis, Chronic Hepatitis, and Control Groups Cirrhosis 7␤-hydroxycholesterol (ng/mL) 7-ketocholesterol (ng/mL) ␣-tocopherol (mg/dL) Oxysterols/␣-tocopherol* Controls Chronic Hepatitis All n ⫽ 32 MELD ⬍ 20† n ⫽ 26 MELD ⱕ 20 n⫽6 3.06 ⫾ 1.78 6.77 ⫾ 3.08 1.28 ⫾ 0.38 8.39 ⫾ 6.25 5.43 ⫾ 4.16¶ 9.25 ⫾ 5.12¶ 0.93 ⫾ 0.36¶ 18.99 ⫾ 16.17¶ 16.93 ⫾ 20.79‡§ 16.93 ⫾ 12.07‡§ 0.75 ⫾ 0.30§ 48.32 ⫾ 34.33‡§ 17.11 ⫾ 22.64 15.80 ⫾ 12.63 0.75 ⫾ 0.30 46.70 ⫾ 35.49 16.15 ⫾ 10.75 20.68 ⫾ 10.01 0.78 ⫾ 0.34 54.25 ⫾ 31.94 * (7␤-hydroxycholesterol ⫹ 7-ketocholesterol)/␣-tocopherol. † See Northup et al.3 ‡ P ⬍ 0.005 vs. chronic hepatitis. § P ⬍ 0.00001 vs. controls. ¶ P ⬍ 0.005 vs. controls. 1501 7␤-hydroxycholesterol and IPGF Table 5. 7␤-hydroxycholesterol, 7-ketocholesterol, and ␣-tocopherol Concentrations and Oxysterol to ␣-tocopherol Ratio of Liver With Cirrhosis and Normal Liver ␣-tocopherol (␮g/g protein) ␣-tocopherol (␮g/g liver) ␣-tocopherol (mg/g cholesterol) 7␤-hydroxycholesterol (␮g/g protein) 7-ketocholesterol (␮g/g protein) 7␤-hydroxycholesterol (␮g/g liver) 7-ketocholesterol (␮g/g liver) 7␤-hydroxycholesterol (mg/g cholesterol) 7-ketocholesterol (mg/g cholesterol) oxysterols/␣-tocopherol* Liver With Cirrhosis n ⫽ 21 Normal Liver n ⫽ 19 P 34.69 ⫾ 51.83 2.35 ⫾ 3.48 1.12 ⫾ 1.49 1.58 ⫾ 0.99 4.03 ⫾ 6.81 0.12 ⫾ 0.13 0.29 ⫾ 0.47 0.06 ⫾ 0.06 0.13 ⫾ 0.12 1.65 ⫾ 2.59 72.09 ⫾ 93.28 6.12 ⫾ 7.06 3.85 ⫾ 5.19 0.74 ⫾ 0.40 2.38 ⫾ 1.41 0.07 ⫾ 0.03 0.24 ⫾ 0.11 0.04 ⫾ 0.04 0.18 ⫾ 0.15 0.17 ⫾ 0.21 0.027 0.014 0.011 Ns Ns Ns Ns Ns Ns 0.021 * (7␤-hydroxycholesterol ⫹ 7-ketocholesterol)/␣-tocopherol. Hepatic Tissue Oxysterol and Vitamin E Content in Cirrhotic and Normal Liver Data regarding the balance between oxidative stress and antioxidant defense in liver specimens are shown in Table 5. Livers with cirrhosis showed significantly lower hepatic ␣-tocopherol content compared with normal livers. This difference existed whether ␣-tocopherol was expressed as a function of either tissue weight, cholesterol, or hepatic protein content. No difference in the hepatic content of oxysterols was observed between livers with cirrhosis and normal livers. The oxysterol/␣-tocopherol ratio in liver tissue was significantly higher in livers with cirrhosis compared with normal livers. Discussion The main and novel finding of the present study is that pretransplantation recipient plasma 7␤-hydroxycholesterol, a marker of systemic oxidative stress and lipid peroxidation, is associated with the quality of early graft function after adult, non-status 1, full-size, primary OLT performed with deceased donors. At univariate analysis, relatively high recipient plasma 7␤-hydroxycholesterol concentration and, in agreement with a recent report from our group, low donor preharvest PaO2 were significantly associated with poor early postoperative graft function.2 In the present study, we also found a trend for a more frequent use of noradrenaline in the donor and for a higher recipient MELD score. At multivariate analysis, the recipient plasma 7␤-hydroxycholesterol level remained independently associated with early graft function even after controlling for recipient MELD score, donor history of cardiac arrest and cause of death, and other parameters demonstrated to predict the quality of early graft function and survival in previous studies, like graft cold ischemia time, donor age, serum sodium concentration, length of intensive care stay, noradrenaline infusion, and preharvest PaO2.1-3 In particular, for every 1-ng/mL increase in plasma 7␤-hydroxycholesterol, odds of IPGF increased by 17%. The fact that we found such a strong association between plasma 7␤-hydroxycholesterol and IPGF in a relatively small number of transplants strengthens its power. If our results will be confirmed in larger series, plasma 7␤-hydroxycholesterol concentration might be a recipient variable relevant to transplant allocation policy. On the other hand, the relatively small number of transplants in the present study could have caused a ␤ type error masking the association with the quality of graft function of the other above-mentioned variables described in larger series.1-3 It is also likely that the effect of some of the variables shown to predict the quality of graft function in transplants performed a few years ago, like graft cold ischemia time, donor serum sodium concentration and length of intensive care stay, has been blunted in the present study by changes in surgical and intensive care unit management in the last years based upon previous studies (i.e., shorter graft cold ischemia times, hypernatriemia treatment, allocation policy minimizing the simultaneous occurrence of negative cofactors). Markers of oxidative stress such as plasma malondialdehyde (MDA) lack specificity and sensitivity.26,32 MDA assays, albeit useful in in vitro systems, are unsatisfactory and give contrasting results when applied in vivo. In fact, increased blood levels of MDA were found in HCV chronic liver disease C by Yadav et al.21 but not 1502 Corradini et al. by Jain et al.,16 who, however, found increased levels of urinary isoprostanes, the gold standard for the assessment of systemic oxidative injury in vivo.33 In the present study, we used a specific mass spectrometric assay of plasma 7␤-hydroxycholesterol and 7-ketocholesterol, validated markers of lipid peroxidation in vivo which have been previously demonstrated as being elevated in several clinical conditions characterized by high systemic oxidative stress status.26,34,35 In addition, we simultaneously evaluated oxysterols and ␣-tocopherol, obtaining a more detailed picture of the oxidant/ antioxidant balance in the same plasma sample of each patient. In agreement with previous studies on urinary and plasma isoprostanes16,22 and on plasma ␣-tocopherol,19,36 we found a deterioration of the constitutive antioxidant ␣-tocopherol that paralleled the increase in oxysterol concentration in plasma of patients with chronic hepatitis and in a more pronounced way in those with cirrhosis, as compared to healthy controls, suggesting evidence of increased radical flux in vivo as the natural history of chronic liver disease progresses. These findings are consistent with the kinetics of lipid peroxidation in vitro during which the evolution of oxidation products is preceded by consumption of constitutive antioxidants.37 In the present study, no causative relation can be demonstrated between high preoperative plasma 7␤-hydroxycholesterol levels and IPGF. However, we failed to demonstrate differences in plasma 7␤-hydroxycholesterol levels of patients according to the severity of cirrhosis expressed by the MELD score. If this finding will be confirmed in larger series, the association of high plasma 7␤-hydroxycholesterol and IPGF could be explained by the oxysterol biological activities rather than merely reflect more advanced liver disease and could support attempts to reduce plasma oxysterol concentration with antioxidant supplementation in patients undergoing OLT.26 In keeping with this hypothesis, plasma oxysterols might be potentially toxic at physiological concentrations, in analogy to their injurious effect on arterial endothelium in atherosclerosis, on sinusoidal endothelial cells whose damage is a key step in the hepatic ischemia-reperfusion injury.1,4,28,38-40 In the present study, we tried to indirectly identify the site of enhanced free radical flux in cirrhosis in vivo by comparing the oxidative stress/antioxidant status at the plasma and the hepatic tissue level in cirrhosis, at tissue level in the liver with cirrhosis and the normal liver, and at the plasma level in patients with cirrhosis before and 1 month after OLT. In patients with cirrhosis we observed a different distribution in oxysterols and ␣-tocopherol in the liver compared with the plasma compartment. In plasma we found a reduction of ␣-tocopherol and an increase in oxysterol concentration in patients with cirrhosis compared with healthy patients. In liver specimens, we found a reduction of ␣-tocopherol, but we did not detect any difference in oxysterol concentration in organs with cirrhosis compared with normal organs. We hypothesize that a high flux of free radicals occurs in the liver tissue, leading to ␣-tocopherol consumption and to production of oxysterols that are not retained in the liver but are released and accumulated in the plasma compartment. This hypothesis is supported by a recent report demonstrating that experimental cholestatic liver disease is associated with increased lipid peroxidation in plasma, kidney, heart, and brain.41 However, it is possible that the source of free radicals in chronic liver disease is also systemic and the liver acts as a target of free radical damage. In support of the latter concept, we observed that the unbalanced systemic oxidative stress/antioxidant status present in patients with cirrhosis was only partially corrected in the short term after OLT. In fact, after 1 month of follow-up, plasma ␣-tocopherol significantly increased, reaching values comparable to healthy subjects, but oxysterol concentrations remained elevated. The recovery of normal ␣-tocopherol plasma levels might reflect the re-establishment of function in the liver, which is crucial for ␣-tocopherol metabolism and its delivery to peripheral cells.42 On the other hand, the persistence of high oxysterol plasma levels 1 month after OLT should be independent from free radical production in the liver by ischemia-reperfusion injury. In keeping with this hypothesis, two previous studies evaluated the timing of systemic oxidative stress changes, as expressed by plasma MDA and urinary isoprostanes, in the post-OLT period.24,25 These studies showed an immediate and transient increase of plasma oxidative stress markers after OLT as expression of reperfusion injury, followed by a rapid reduction to baseline or lower values that actually remained elevated compared with healthy subjects during the 20 days and 12 months of follow-up.24,25 Different in vivo kinetics of each specific oxidative stress marker studied could account for the fact that, 20 to 30 days after OLT, oxysterols in the present study, and MDA in Blasi’s study did not differ from baseline, while urinary isoprostanes were significantly reduced compared with the preoperative values in Burke’s study.24,25 The reason for the persistence of increased in vivo systemic radical flux in liver disease patients 1 month after OLT remains at the moment speculative, and estimation of oxysterol kinetics in vivo, particularly in patients with impaired liver function, by 7␤-hydroxycholesterol and IPGF the use of deuterium-labeled compounds is highly desirable. However, our data that show a trend for a more pronounced increase in plasma oxysterols in the posttransplantation phase in HCV-positive compared with HCV-negative subjects allow us to hypothesize that the HCV infection itself can contribute to increase oxidative stress. In fact, HCV infection has been demonstrated to occur in extra-hepatic sites, including mononuclear cells and neutrophils of peripheral blood, and viral replication starts early after liver transplantation.43,44 In addition, it has been demonstrated that the expression of HCV core protein in different cell lines is associated with increased free radical production by possible delocalization of the mithocondrial electron transport chain, and Huh-7 cell lines transfected with HCV NS5A expression vector show increased free radical flux in mitochondria.45,46 Taken together, our preliminary data show that a systemic imbalance of oxidative stress/antioxidant status is progressively present in the natural history of chronic liver disease, without a further worsening once the cirrhotic stage has been reached. In addition, we found an association between increased recipient systemic radical flux and IPGF after OLT. If confirmed in larger series, our findings could have important clinical implications for the prevention of IPGF. Acknowledgment This work was supported by grants from Associazione Italiana Ricerca sul Cancro, Fondazione Cassa Risparmio di Roma, and MIUR (COFIN 2002058281 003) to IL, and from the Swedish Heart Lung Foundation to U.D. References 1. Busuttil RW, Tanaka K. The utility of marginal donors in liver transplantation. Liver Transpl 2003;651-653. 2. Ginanni Corradini S, Elisei W, De Marco R, Siciliano M, Iappelli M, Pugliese F,et al. Preharvest donor hyperoxia predicts good early graft function and longer graft survival after liver transplantation. Liver Transpl 2005;11:140-151. 3. Northup PG, Berg CL. Preoperative delta-MELD score does not independently predict mortality after liver transplantation. Am J Transplant 2004;4:1643-1649. 4. Selzner N, Rudiger H, Graf R, Clavien PA. Protective strategies against ischemic injury of the liver. Gastroenterology 2003;125: 917-936. 5. Omar R, Nomikos I, Piccorelli G, Savino J, Agarwal N. Prevention of postischaemic lipid peroxidation and liver cell injury by iron chelation. Gut 1989;30:510-514. 6. Mizunuma K, Ohdan H, Tashiro H, Fudaba Y, Ito H, Asahara T. Prevention of ischemia-reperfusion-induced hepatic microcirculatory disruption by inhibiting stellate cell contraction using rock inhibitor. Transplantation 2003;75:579-586. 1503 7. Poli G. Pathogenesis of liver fibrosis: role of oxidative stress. Mol Aspects Med 2000;21:49-98. 8. Parola M, Robino G. Oxidative stress-related molecules and liver fibrosis. J Hepatol 2001;35:297-306. 9. Jaeschke H. Reactive oxygen and mechanisms of inflammatory liver injury. J Gastroenterol Hepatol 2000;15:718-724. 10. Tsukamoto H, Rippe R, Niemela O, Lin M. Roles of oxidative stress in activation of Kupffer and Ito cells in liver fibrogenesis. J Gastroenterol Hepatol 1995;10:S50 – S53. 11. Arteel GE. Oxidants and antioxidants in alcohol-induced liver disease. Gastroenterology 2003;124:778-790. 12. Loguercio C, Federico A. Oxidative stress in viral and alcoholic hepatitis. Free Radic Biol Med 2003;34:1-10. 13. Namikawa C, Shu-Ping Z, Vyselaar JR, Nozaki Y, Nemoto Y, Ono M, et al. Polymorphisms of microsomal triglyceride transfer protein gene and manganese superoxide dismutase gene in nonalcoholic steatohepatitis. J Hepatol 2004;40:781-786. 14. Aboutwerat A, Pemberton PW, Smith A, Burrows PC, McMahon RF, Jain SK, et al. Oxidant stress is a significant feature of primary biliary cirrhosis. Biochim Biophys Acta 2003;1637:142150. 15. Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115: 209-218. 16. Jain SK, Pemberton PW, Smith A, McMahon RF, Burrows PC, Aboutwerat A, et al. Oxidative stress in chronic hepatitis C: not just a feature of late stage disease. J Hepatol 2002;36:805-811. 17. Higueras V, Raya A, Rodrigo JM, Serra MA, Roma J, Romero FJ. Interferon decreases serum lipid peroxidation products of hepatitis C patients. Free Radic Biol Med 1994;16:131-133. 18. Ferre N, Camps J, Prats E, Girona J, Gomez F, Heras M, et al. Impaired vitamin E status in patients with parenchymal liver cirrhosis: relationships with lipoprotein compositional alterations, nutritional factors, and oxidative susceptibility of plasma. Metabolism 2002;51:609-615. 19. Nagita A, Ando M. Assessment of hepatic vitamin E status in adult patients with liver disease. Hepatology 1997;26:392-397. 20. Boya P, de la Pena A, Beloqui O, Larrea E, Conchillo M, Castelruiz Y, et al. Antioxidant status and glutathione metabolism in peripheral blood mononuclear cells from patients with chronic hepatitis C. J Hepatol 1999;31:808-814. 21. Yadav D, Hertan HI, Schweitzer P, Norkus EP, Pitchumoni CS. Serum and liver micronutrient antioxidants and serum oxidative stress in patients with chronic hepatitis C. Am J Gastroenterol 2002;97:2634-2639. 22. Pratico D, Iuliano L, Basili S, Ferro D, Camastra C, Cordova C, et al. Enhanced lipid peroxidation in hepatic cirrhosis. J Invest Med 1998;46:51-57. 23. Meagher EA, Barry OP, Burke A, Lucey MR, Lawson JA, Rokach J, et al. Alcohol-induced generation of lipid peroxidation products in humans. J Clin Invest 1999;104:805-813. 24. Biasi F, Bosco M, Chiappino I, Chiarpotto E, Lanfranco G, Ottobrelli A, et al. Oxidative damage in human liver transplantation. Free Radic Biol Med 1995;19:311-317. 25. Burke A, FitzGerald GA, Lucey MR. A prospective analysis of oxidative stress and liver transplantation. Transplantation 2002; 74:217-221. 26. Micheletta F, Natoli S, Misuraca M, Sbarigia E, Diczfalusy U, Iuliano L. Vitamin E supplementation in patients with carotid atherosclerosis: reversal of altered oxidative stress status in plasma but not in plaque. Arterioscler Thromb Vasc Biol 2004;24:136140. 27. Iuliano L, Micheletta F, Natoli S, Ginanni Corradini S, Iappelli 1504 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. Corradini et al. M, Elisei W, et al. Measurement of oxysterols and alpha-tocopherol in plasma and tissue samples as indices of oxidant stress status. Anal Biochem 2003;312:217-223. Zhou Q, Wasowicz E, Handler B, Fleischer L, Kummerow FA. An excess concentration of oxysterols in the plasma is cytotoxic to cultured endothelial cells. Atherosclerosis 2000;149:191-197. Gonzalez FX, Rimola A, Grande L, Antolin M, Garcia-Valdecasas JC, Fuster J, et al. Predictive factors of early postoperative graft function in human liver transplantation. Hepatology 1994; 20:565-573. Chui AKK, Haghighi K, Painter D, Jayasundera M, Hall G, Rao ARN, et al. Donor fatty (steatotic) liver allograft in orthotopic liver transplantation. Transplant Proc 1998;30:3286-3287. Nanashima A, Pillay P, Verran DJ, Painter D, Nakasuji M, Crawford M, et al. Analysis of initial poor graft function after orthotopic liver transplantation: experience of an Australian single liver transplantation center. Transplant Proc 2002;34:12311235. Griendling KK, FitzGerald GA. Oxidative stress and cardiovascular injury: Part I: basic mechanisms and in vivo monitoring of ROS. Circulation 2003;108:1912-1916. Moore K. Isoprostanes and the liver. Chem Phys Lipids 2004; 128:125-133. Mol MJ, de Rijke YB, Demacker PN, Stalenhoef AF. Plasma levels of lipid and cholesterol oxidation products and cytokines in diabetes mellitus and cigarette smoking: effects of vitamin E treatment. Atherosclerosis 1997;129:169-176. Salonen JT, Nyyssonen K, Salonen R, Porkkala-Sarataho E, Tuomainen TP, Diczfalusy U, et al. Lipoprotein oxidation and progression of carotid atherosclerosis. Circulation 1997;95:840845. Look MP, Reichel C, von Falkenhausen M, Hahn C, Stockinger K, von Bergmann K, et al. Vitamin E status in patients with liver cirrhosis: normal or deficient? Metabolism 1999;48:86-91. Patel RP, Diczfalusy U, Dzeletovic S, Wilson MT, Darley-Usmar VM. Formation of oxysterols during oxidation of low den- 38. 39. 40. 41. 42. 43. 44. 45. 46. sity lipoprotein by peroxynitrite, myoglobin, and copper. J Lipid Res 1996;37:2361-2371. Rusinol AE, Thewke D, Liu J, Freeman N, Panini SR, Sinensky MS. AKT/protein kinase B regulation of BCL family members during oxysterol-induced apoptosis. J Biol Chem 2004;279: 1392-1399. Pirillo A, Zhu W, Roma P, Galli G, Caruso D, Pellegatta F, et al. Oxysterols from oxidized LDL are cytotoxic but fail to induce hsp70 expression in endothelial cells. FEBS Lett 1999;462:113116. Meynier A, Andre A, Lherminier J, Grandgirard A, Demaison L. Dietary oxysterols induce in vivo toxicity of coronary endothelial and smooth muscle cells. Eur J Nutr. 2005;Jan 27:[Epub ahead of print]. Ljubuncic P, Tanne Z, Bomzon A. Evidence of a systemic phenomenon for oxidative stress in cholestatic liver disease. Gut 2000;47:710-716. Stocker A, Azzi A. Tocopherol-binding proteins: their function and physiological significance. Antioxid Redox Signal 2000;2: 397-404. Crovatto M, Pozzato G, Zorat F, Pussini E, Nascimben F, Baracetti S, et al. Peripheral blood neutrophils from hepatitis C virusinfected patients are replication sites of the virus. Haematologica 2000;85:356-361. Garcia-Retortillo M, Forns X, Feliu A, Moitinho E, Costa J, Navasa M, et al. Hepatitis C virus kinetics during and immediately after liver transplantation. Hepatology 2002;35: 680-687. Okuda M, Li K, Beard MR, Showalter LA, Scholle F, Lemon SM, et al. Mitochondrial injury, oxidative stress, and antioxidant gene expression are induced by hepatitis C virus core protein. Gastroenterology 2002;122:366-375. Gong G, Waris G, Tanveer R, Siddiqui A. Human hepatitis C virus NS5A protein alters intracellular calcium levels, induces oxidative stress, and activates STAT-3 and NF-kappa B. Proc Natl Acad Sci U S A 2001;98:9599-9604.