Telephone: +34957010132; FAX: +34957012876
Impact of donor graft steatosis on overall outcome and viral recurrence after liver transplantation for hepatitis C virus cirrhosis†
Article first published online: 24 DEC 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 15, Issue 1, pages 37–48, January 2009
How to Cite
Briceño, J., Ciria, R., Pleguezuelo, M., de la Mata, M., Muntané, J., Naranjo, Á., Sánchez-Hidalgo, J., Marchal, T., Rufián, S. and López-Cillero, P. (2009), Impact of donor graft steatosis on overall outcome and viral recurrence after liver transplantation for hepatitis C virus cirrhosis. Liver Transpl, 15: 37–48. doi: 10.1002/lt.21566
See Editorial on Page 4
- Issue published online: 24 DEC 2008
- Article first published online: 24 DEC 2008
- Manuscript Accepted: 16 MAY 2008
- Manuscript Received: 18 FEB 2008
- Centro de Investigación Biomédica en Red en Enfermedades Hepáticas y Digestivas
The aim of this study was to determine the influence of donor graft steatosis on overall outcome, viral recurrence, and fibrosis progression in orthotopic liver transplantation (OLT) for hepatitis C virus (HCV) cirrhosis. One hundred twenty patients who underwent OLT for HCV cirrhosis between 1995 and 2005 were included in the study. Donor steatosis was categorized as absent (0%-10%; n = 40), mild (10%-30%; n = 32), moderate (30%-60%; n = 29), or severe (>60%; n = 19). A Cox multivariate analysis for marginal donor variables and a Model for End-Stage Liver Disease index were performed. Fibrosis evolution was analyzed in liver biopsies (fibrosis < 2 or ≥2) 3, 6, and 12 months post-OLT and in the late post-OLT period. Fifty-six grafts were lost (46%). The survival of the grafts was inversely proportional to donor liver steatosis: 82%, 72%, and 72% at 1, 2, and 3 years post-OLT in the absence of steatosis; 73%, 63%, and 58% with mild steatosis; 74%, 62%, and 43% with moderate steatosis; and 62%, 49%, and 42% with severe steatosis (P = 0.012). HCV recurrence was earlier and more frequent in recipients with steatosis > 30% (46% versus 32% at 3 months, P = 0.017; 58% versus 43% at 6 months, P = 0.020; 70% versus 56% at 12 months, P = 0.058; and 95% versus 69% at 3 years post-OLT, P = 0.0001). Graft survival was lower in alcoholic liver disease recipients versus HCV recipients when steatosis was >30% at 3, 6, and 12 months post-OLT (P = 0.042) but not when steatosis was <30% (P = 0.53). A higher fibrosis score was obtained 3 months post-OLT (P = 0.033), 6 months post-OLT (P = 0.306), 12 months post-OLT (P = 0.035), and in the late post-OLT period (P = 0.009). In conclusion, donor graft steatosis influences the outcome of OLT for HCV cirrhosis. HCV recurrence is more frequent and earlier in recipients of moderately and severely steatotic livers. Fibrosis evolution is higher when graft steatosis is >30%. OLT with >30% steatotic donor livers should be precluded in HCV recipients. Liver Transpl 15:37–48, 2009. © 2008 AASLD.
The World Health Organization has reported that 3% of patients are infected with hepatitis C virus (HCV); this represents 170 million people worldwide.1, 2 HCV is the most common cause of chronic liver diseases and hepatocellular carcinoma (HCC), and there is important geographic variability.3, 4 Moreover, HCV has been identified as the major indication for orthotopic liver transplantation (OLT) in Western countries.5, 6
Although differences in patient or graft survival among HCV and non-HCV patients who have undergone OLT have not been described,7–9 some studies have recently reported a reduction in the long-term survival rates in HCV-infected liver transplant recipients.10 Viral recurrence is now considered universal,11 and most patients develop histological hepatitis, with progression to cirrhosis in up to 30%, after 5 years.12
Many factors have been identified as potential risk factors responsible for the progressively decreasing survival rates. The age and race of the recipient, presence of HCC in the explanted organ,13, 14 pretransplant and posttransplant viral load, and viral genotype,15, 16 as well as the immunosuppressive protocol, acute rejection episodes, and treatment for HCV recurrence,17–19 are factors related to bad prognosis in HCV transplanted patients. Unfortunately, donor factors influencing the outcome of posttransplant HCV recipients have not been widely studied. Actually, the donor age, the immunosuppressant regimen, the number of episodes of rejection, and a few findings about living donor liver transplantation seem to be the only consistent findings that adversely affect graft and patient survival in HCV posttransplant recipients.20, 21 HCV rapidly infects the graft after liver transplantation, and clinical or histological markers of viral recurrence can be seen in 40% to 60% of the patients. Finally, 5% to 10% will require retransplantation because of graft failure secondary to cirrhosis or a severe form of cholestatic hepatitis.22, 23
This study evaluates the influence of some marginal donor factors in the rapid progression of viral recurrence and liver cirrhosis in HCV transplanted patients and investigates mainly how the degree of donor graft steatosis influences liver preservation injury after transplantation.24 The aims of the present study are (1) to examine the influence of donor graft steatosis on overall patient survival and graft outcome in liver transplantation for HCV cirrhosis, (2) to measure the effect that the combination of marginal donor variables with graft steatosis has on patient and graft survival, (3) to analyze the frequency of viral recurrence with increasing values of donor graft steatosis in HCV recipients, (4) to determine the timing of viral recurrence with different grades of liver fat content, and (5) to analyze the severity of viral recurrence and outcome in patients receiving steatotic donor grafts.
PATIENTS AND METHODS
A retrospective analysis was performed that included 487 primary liver transplants in 450 patients, 64 of which were pediatric, for end-stage liver cirrhosis from January 1995 to December 2005 at Reina Sofia University Hospital.
One hundred seventy-five (36%) transplants were performed in HCV recipients. Fifty-five (31.4%) HCV patients with HCC as the main indication for OLT were excluded from our study to avoid interference with the relative contributions of tumor and viral-related factors to the outcome of OLT. HCV patients with incidental HCC (n = 6; 5%) were included in our study because the prognosis of OLT with incidental HCC is not different from that of HCV alone.25 One hundred twenty transplants performed in 106 HCV recipients were finally included. Four patients (3.8%) were coinfected with hepatitis B virus, 21 (19.8%) had a history of significant alcohol intake, and 2 (1.8%) had coexisting hemochromatosis. Fifteen (12.5%) retransplants were performed on 14 patients; 1 of these patients required a third liver retransplantation intervention.
All patients signed informed consent forms allowing sample collection before undergoing OLT. This study was reviewed and approved by the Department of Ethics in Biomedical Research of Reina Sofia University Hospital.
HCV Detection and HCV-RNA Measurements
The diagnosis of HCV infection before OLT was based on the detection of anti-HCV seropositivity with an enzyme-linked immunoabsorbent assay (ELISA 2.0). Viral evidence from a polymerase chain reaction test for HCV-RNA and qualitative tests for the viral genotype have been routinely applied in our center since 2001. Accordingly, the values for HCV-RNA levels in serum and the genotype are not reported.
Donor Variables and Graft Steatosis Assessment
A liver biopsy from the native explanted liver was obtained for all subjects for histological analysis. An intraoperative liver biopsy was obtained for the analysis of ischemia-reperfusion injury and steatosis in the donor liver before implantation but with cold preservation following standard methods, and another was obtained after complete revascularization.
Hepatic macrovesicular steatosis was quantified with the previously validated scale proposed by Adinolfi et al.,26 which stratifies hepatic steatosis into 4 categories: grade 1 (0%-10% steatosis), grade 2 (11%-30% steatosis), grade 3 (31%-60% steatosis), and grade 4 (>60% steatosis). According to previous reports, a statistical analysis of the data was also performed with the distribution of the <30% steatotic grafts and >30% steatotic grafts (≥30%; n = 25, 20.8%).27, 28 Because microvesicular steatosis is generally disregarded as a result of low clinical significance, all grade designations were based only on macrovesicular steatosis.
The histological features used to assess the preservation injury were the severity, type, and location of necrosis and inflammation as well as the location and severity of hepatocellular swelling and cytoaggregation.29, 30 It was accordingly divided into 4 grades, which are described in Table 1.24
|Grade of Preservation Injury||Grade I (Minimal)||Grade II (Mild)||Grade III (Moderate)||Grade IV (Severe)|
|Severity of inflammation (number of inflammatory cells per high power field)||0–5||6–10||11–20||>20|
|Severity of necrosis, cytoaggregation, and hepatocyte swelling (percentage of cells)||<10%||10%–40%||40%–70%||>70%|
The following donor variables with the corresponding cutoff point values were also examined to assess their influence on the steatotic graft liver function: gender, donor age (≥55 years old; n = 30, 25%), prolonged intensive care unit stay under ventilatory support (≥4 days; n = 30, 25%), extended cold ischemia time (≥12 hours; n = 23, 19.16%), high inotropic drug use (n = 50, 41.66%) including high dopamine doses (≥15 μg/kg/minute; n = 13, 10.83%) or epinephrine, norepinephrine, or dobutamine at any dose (n = 37, 30.83%), hypotensive episodes (>1 hour and <60 mm Hg systolic blood pressure; n = 34, 28.3%), plasma sodium level (≥155 mEq/L; n = 17, 14.16%), bilirubin (≥2 mg/dL; n = 6, 5%), alanine aminotransferase (ALT; ≥200 U/L; n = 5, 4.16%), and aspartate aminotransferase (AST; ≥200 U/L; n = 8, 6.6%).24
Patients who were undergoing oral hypoglycemic therapy and/or had an insulin requirement before OLT were diagnosed as having pre-OLT diabetes mellitus. New-onset post-OLT diabetes mellitus was defined as de novo and persistent hyperglycemia requiring long-term treatment with oral hypoglycemic agents and/or insulin.
Donor Procurement and Surgical Technique
Liver retrieval was carried out via aortic and portal perfusion with University of Wisconsin solution.31, 32 A non–heart-beating donor program has still not been achieved in our center, so no graft of that type was obtained. The transplantation procedures were performed with the piggyback technique with preservation of the full length of the recipient inferior vena cava and choledochocholedochostomy.33, 34
A triple therapy with cyclosporine A (CsA) or tacrolimus (Tac), azathioprine, and prednisolone was the initial immunosuppressive regimen. The most commonly used regimen consisted of a triple CsA-based induction therapy including CsA, azathioprine/mycophenolate mofetil, and steroids. Tac-based regimens were dual (Tac and steroids) or triple (Tac, azathioprine/mycophenolate mofetil, and steroids). Acute cellular rejections were scored on the basis of the Banff rejection activity index.35 All of them were confirmed by histologic analysis and treated with methylprednisolone (0.5-1 g/day for 3 days). All patients who were under treatment for rejection also received ganciclovir for cytomegalovirus prophylaxis.
The recurrence of HCV disease was diagnosed by the presence of biochemical markers of graft dysfunction with concomitant liver biopsy showing histological features related to recurrent HCV. The time of recurrence was divided into 4 stages: up to 3 months post-transplant, from 3 to 6 months post-transplant, from 6 to 12 months post-transplant, and over 1 year post-transplant. This period comprises the time between OLT and the first biopsy showing HCV recurrence according to the criteria described previously and is independent of the severity of that recurrence.
Histological Assessment of Viral Recurrence
All histological examinations were performed by an expert liver pathologist (T.M.) who evaluated all explanted and liver biopsy specimens in a blinded fashion. The major portion of each biopsy specimen was fixed in 10% neutral buffered formalin, and the sections were routinely stained with hematoxylin-eosin, reticulin, and Perl's and Orcein's stains. A few biopsies were stained with Masson-Goldner, iron, and a periodic Schiff reagent. A small portion of the biopsy specimen was embedded in Epon-Araldite for transmission electron microscopy.
A percutaneous posttransplant liver biopsy was obtained only when clinically relevant elevations of aminotransferases (1.5 times normal values) or other signs of graft dysfunction (bilirubin ≥ 2 mg/dL) were found. Histological recurrence was diagnosed and quantified according to a slight modification of the histological classification proposed by Scheuer et al.36 Accordingly, a scale of 0 to 3 was used to define the degree of fibrosis (0: absent; 1: mild portal fibrosis and/or mild central fibrosis without septa; 2: moderate fibrosis with few septa; 3: numerous septa or cirrhosis).
All data were collected in a retrospective database. Data are presented as the mean ± standard deviation with an indication of the interval range for statistical significance, which was fixed at P < 0.05. The characteristics of the patients were compared by the chi-square test for categorical variables (Fisher's test was applied when necessary) and the Student t test, Mann-Whitney U test, and analysis of variance for continuous variables. Univariate data comparison among groups was performed with life tables with the Kaplan-Meier method and the log-rank test. The Cox proportional hazard regression model was used for multivariate analysis under the assumption of statistically significant values under P ≤ 0.1. All analyses were performed with SPSS 10.0 (SPSS, Inc., Chicago, IL).
One hundred twenty transplanted patients were included in our study, 75 males (62.5%) and 45 females (37.5%), with a mean age of 51 years (range = 31-66). A summary of demographic/clinical characteristics of the donors and recipients is displayed in Table 2.
|Total number of patients/transplants||106/120|
|Gender male||75/120 (62.5%)|
|Gender female||45/120 (37.5%)|
|Recipient age||51.45 ± 8.23 (31–66)|
|Pretransplant INR||1.59 ± 0.40 (1.00–2.80)|
|Pretransplant serum creatinine (mg/dL)||1.32 ± 1.21 (0.30–8.90)|
|Pretransplant total bilirubin (mg/dL)||5.83 ± 6.74 (0.50–38.90)|
|Pretransplant MELD||17.53 ± 7.14 (5.00–46.00)|
|Gender male||77/120 (64.2%)|
|Gender female||43/120 (35.8%)|
|Donor age||39.68 ± 17.55 (9–77)|
|≥55 years||30/120 (25%)|
|ITU stay (days)||2.74 ± 2.87 (0.50–14.00)|
|≥4 days||30/120 (25%)|
|Cold ischemia time (minutes)||479.88 ± 191.02 (35–900)|
|≥12 hours||23/120 (19.16%)|
|Dose (μg/kg/minute)||5.90 ± 6.97 (0–50)|
|>15 (μg/kg/minute)||13/120 (10.83%)|
|Epinephrine or norepinephrine or dobutamine||37/120 (30.83)|
|Donor hypotensive episodes||34/120 (28.23%)|
|Donor plasma sodium levels||145.54 ± 8.88 (126–171)|
|>155 mEq/L||17 (14.16%)|
|Bilirubin (mg/dL)||0.89 ± 0.61 (0.12–4.50)|
|≥2 mg/dL||6/120 (5%)|
|ALT (U/L)||57.92 ± 131.67 (4.00–1244)|
|≥200 U/L||5/120 (4.16%)|
|AST (U/L)||67.26 ± 112.16 (10.0–730)|
|≥200 U/L||8/120 (6.6%)|
Descriptive Analysis of the Baseline Marginal Donor and Recipient Variables According to Graft Steatosis
Four groups of steatotic grafts were identified according to previous reports.26–28 Only recipient gender showed statistically significant differences in the analysis of the distribution of marginal variables (P = 0.029). Thus, the male gender distribution was 18 (15%) in the no–steatotic liver group, 20 (16.7%) in the mildly steatotic group, 23 (19.2%) in the moderately steatotic group, and 14 (11.7%) in the severely steatotic group. On the other hand, the female gender distribution was 21 (17.5%), 12 (10%), 6 (5%), and 5 (4.16%), respectively, in the same groups.
No other statistically significant differences were obtained in the baseline characteristics for the 4 steatotic graft groups with well-matched donor criteria according to liver steatosis (Table 3).
|Distribution of Marginal Donor Variables||Donor Graft Steatosis|
|0%–10% (No)||10%–30% (Mild)||30%–60% (Moderate)||≥60% (Severe)||P|
|Pretransplant MELD||17.64 (9–36)||17.06 (5–46)||17.76 (9–36)||16.77 (10–31)||0.982|
|Pretransplant MELD < 20||0.783|
|MELD < 20||24 (20%)||20 (16.66%)||16 (13%)||14 (11.66%)|
|MELD ≥20||9 (7.5%)||9 (7.5%)||9 (7.5%)||4 (3.33%)|
|Recipient age||52.82 (39–66)||48.93 (31–62)||52.03 (41–66)||51.89 (35–63)||0.239|
|Male||18 (15%)||20 (16.66%)||23 (19.16%)||14 (11.66%)|
|Female||21 (17.5%)||12 (10%)||6 (5%)||5 (4.16%)|
|Donor age||38.79 (15–77)||37.53 (9–67)||39.24 (13–69)||44.42 (17–67)||0.581|
|<55 years||28 (23.33)||26 (21.66%)||23 (19.16%)||13 (10.83%)|
|≥55 years||11 (9.16%)||6 (5%)||6 (5%)||6 (5%)|
|Male||23 (19.16%)||21 (17.5%)||18 (15%)||15 (12.5%)|
|Female||16 (13.33%)||11 (9.16%)||11 (9.16%)||4 (3.33%)|
|No||4 (3.33%)||2 (1.66%)||0 (0%)||1 (0.83%)|
|Mild||17 (14.16%)||16 (13.33%)||14 (11.66%)||10 (8.3%)|
|Moderate||11 (9.16%)||10 (8.3%)||8 (6.66%)||5 (4.16%)|
|Severe||5 (4.16%)||2 (1.66%)||2 (1.66%)||2 (1.66%)|
|Negative||13 (10.83%)||7 (5.83%)||8 (6.66%)||10 (8.3%)|
|Positive||25 (20.83%)||25 (20.83%)||21 (17.5%)||9 (7.5%)|
|Dopamine dose (μg/kg/minute)||5.74 (0–20)||7.09 (0–50)||7.07 (0–20)||2.47 (0–15)||0.094|
|Negative||26 (21.6%)||20 (16.66%)||19 (15.83%)||13 (10.83%)|
|Positive||11 (9.16%)||10 (8.3%)||10 (8.3%)||6 (5%)|
|Donor sodium levels||145.60 (135–160)||144.87 (133–157)||145.03 (126–171)||147.31 (136–160)||0.797|
|Donor hypotensive episodes||0.274|
|Negative||29 (24.16%)||20 (16.66%)||18 (15%)||16 (13.33%)|
|Positive||9 (7.5%)||11 (9.16%)||11 (9.16%)||3 (2.5%)|
|Donor bilirubin levels (mg/dL)||0.83 (0.12–2.70)||0.82 (0.20–4.50)||0.91 (0.30–2.40)||1.03 (0.30–3)||0.723|
|Donor ALT levels (IU/L)||78 (4–1244)||72.5 (10–538)||33.9 (9–150)||29.84 (9–86)||0.380|
|Donor AST levels (IU/L)||80.97 (13–718)||81.93 (13–730)||53.89 (10–335)||33.77 (10–85)||0.376|
|Cold ischemia time (minutes)||474.21 (240–900)||506.96 (35–900)||466.73 (135–900)||480.55 (270–790)||0.870|
|Cold ischemia time||0.626|
|<12 hours||32 (26.66%)||27 (22.5%)||21 (17.5%)||16 (13.33%)|
|≥12 hours||7 (5.83%)||5 (4.16%)||8 (6.66%)||3 (2.5%)|
|Warm ischemia time (minutes)||44 (30–52)||42 (34–49)||39 (32–48)||41 (34–54)||0.547|
|Donor ITU stay (days)||2.80 (0.50–13)||2.38 (0.50–10)||3.69 (0.50–14)||1.78 (0.50–9)||0.119|
There were no differences in the number of acute rejection episodes between donor graft steatosis groups (chi-square = 0.346, P = 0.556).
The prevalence of pre-OLT diabetes mellitus and post-OLT diabetes mellitus according to previously described criteria was equally distributed between donor graft steatosis groups: 18% for total pre-OLT diabetes mellitus and 41% for post-OLT diabetes mellitus. The distribution among donor graft steatosis groups was 17% for pre-OLT diabetes mellitus and 42% for post-OLT diabetes mellitus in the no–steatotic liver group, 19% and 37% in the mildly steatotic group, 18% and 43% in the moderately steatotic group, and 21% and 39% in the severely steatotic group.
Influence of Marginal Donor Variables on Graft Survival
The Kaplan-Meier 1-, 3-, and 5-year patient survival rates were 83%, 81%, and 70%, respectively. The overall graft survival rates at 1, 3, and 5 years were 76%, 73%, and 58%, respectively.
Kaplan-Meier univariate analysis at 3, 6, and 12 months and 3 years post-OLT revealed the following statistically significant results for these 4 variables: age > 55 years (P = 0.024), donor graft macrosteatosis > 30% (P = 0.012), prolonged cold ischemia time > 12 hours (P = 0.0001), and moderate to severe ischemia-reperfusion injury (P = 0.007). In multivariate analysis (Cox regression model), donor graft macrosteatosis > 30% (P = 0.001, odds ratio = 2.590), cold ischemia time > 12 hours (P = 0.0001, odds ratio = 1.308), and donor age > 55 years (P = 0.048, odds ratio = 3.373) were independent predictors of graft survival.
Contribution of Donor Graft Steatosis to Posttransplant Patient and Graft Survival in HCV Recipients
Patient survival significantly decreased with the degree of fatty content (Fig. 1): 85%, 73%, and 73% at 1, 3, and 5 years, respectively, after transplantation with no steatosis; 77%, 70%, and 66%, respectively, with mild steatosis; 76%, 62%, and 62%, respectively, with moderate steatosis; and 75%, 57%, and 47%, respectively, with severe steatosis (chi-square = 3.189, P = 0.036). When these results in patients receiving <30% or ≥30% steatotic grafts are analyzed, the patient survival results are as follows: 82%, 72%, and 66% at 1, 3, and 5 years, respectively, in patients receiving grafts with <30% steatosis and 75%, 58%, and 47%, respectively, in patients receiving grafts with ≥30% steatosis (chi-square = 2.934, P = 0.008).
Graft survival significantly decreased with the degree of fatty content (Fig. 1): 82%, 72%, and 72% at 1, 2, and 3 years, respectively, after transplantation with no steatosis; 73%, 63%, and 58%, respectively, with mild steatosis; 74%, 62%, and 43%, respectively, with moderate steatosis; and 62%, 49%, and 42%, respectively, with severe steatosis (chi-square = 5.793, P = 0.012).
The evolution of the highest post-OLT AST and ALT peaks was evaluated to determine if there were differences according to donor graft steatosis. The postoperative AST peak (mean = 1123.15, range = 52-9461) was significantly higher when donor graft steatosis was >30% (mean = 1027.31, range = 116-3219) than when it was <30% (mean = 581.68, range = 59-3489; P = 0.031). There were no statistically significant differences in the evolution of the post-OLT ALT peak between the <30% (mean = 826.86, range = 71-6152) and >30% (mean = 1137.75, range = 138-4140) donor graft steatosis groups (P = 0.7)
The viral recurrence at 3, 6, and 12 months and in the late period post-OLT was analyzed according to the previously described criteria. The presence of high donor graft steatosis induced statistically significant differences 3 months post-OLT, 6 months post-OLT, and in the late post-OLT period (P = 0.017, P = 0.020, and P = 0.001, respectively; Fig. 2). A nearly statistically significant difference was obtained 12 months post-OLT (P = 0.058).
Donor graft steatosis showed a progressively increased risk of earlier and more frequent viral recurrence after OLT in HCV recipients (P = 0.042; Fig. 3).
Combination of the Main Ischemia-Reperfusion Injury Factors with Donor Graft Steatosis
Cold ischemia time and liver preservation injury are the main factors defining the severity of ischemia-reperfusion injury. When cold ischemia times of <12 and >12 hours were divided according to donor graft groups with <30% steatosis and ≥30% steatosis, a statistically significant difference could be observed in Kaplan-Meier's univariate analysis (chi-square = 6.432, P = 0.011; Fig. 4). A similar result was observed with liver preservation injury, divided into no-mild injury and moderate-severe injury, between donor grafts with <30% and ≥30% fatty content (chi-square = 1.649, P = 0.019; Fig. 5).
Comparison of HCV and Non-HCV Recipients
Further analysis was performed to compare the influence of the studied variables in the nonviral recipients. One hundred thirty-six consecutively transplanted patients with end-stage alcoholic liver disease (ALD) in the same period were also included for this analysis. Thirty were associated with HCV infection, and 9 of these presented HCC that became the main indication for OLT. Eighteen patients were associated with HCC (1 of them was incidental). Two patients had a coexisting hepatitis B virus infection. In order to meet the same inclusion criteria previously used for HCV recipients, we excluded all of the previous associations except for the patient with incidental HCC. The final number of patients for this group was 87.
Overall graft survival in the group of patients who underwent OLT for ALD and HCV cirrhosis was stratified according to ≥30% or <30% donor graft steatosis. The overall graft survival 3, 6, and 12 months post-OLT was 94%, 91%, and 88%, respectively, for ALD patients and 90%, 87%, and 80% for HCV-cirrhosis patients with <30% donor graft steatosis (P = 0.53). The overall survival 3, 6, and 12 months post-OLT was 85%, 84%, and 81%, respectively, for ALD patients and 78%, 76%, and 63% for HCV-cirrhosis patients with ≥30% donor graft steatosis (P = 0.042). These results clearly show the higher impact of donor steatosis on graft survival in HCV recipients versus non-HCV recipients (Fig. 6).
Donor Graft Steatosis Increases the Severity of Post-OLT HCV Recurrence
The analysis of the influence of donor graft steatosis as the main donor factor for viral recurrence when marginal donors are used is displayed in Fig. 7. A statistically significant decrease in patient survival was registered when graft steatosis was ≥30%. In this sense, no significant differences in patient survival were observed when viral recurrence did not appear in the post-OLT period between groups with <30% or ≥30% steatosis. However, when viral recurrence was present, the influence of receiving a graft with a fatty content ≥ 30% implied a decrease in overall patient survival by Kaplan-Meier's log rank test (chi-square = 2.917, P = 0.008).
Liver fibrosis is the main indicator used to measure the severity of viral recurrence on the basis of our experience and the absence of a viral load test in the first years of this study. The patients were distributed according to the degree of liver fibrosis (<2 or ≥2) proposed by Scheuer et al.36 An analysis of liver fibrosis according to different groups receiving grafts with <30% or ≥30% steatosis was performed 3, 6, and 12 months post-OLT and in a late post-OLT period (Fig. 8). A comparison with the Bonferroni correction was performed according to the number of recurrences in each stage.
All the patients who developed a histologically demonstrable recurrence at 3 months post-OLT and who received a graft with <30% steatosis had a mild histological recurrence (Fibrosis 0/1). However, patients receiving a graft with ≥30% steatosis were distributed according to <2 (87%) and ≥2 (13%; 1 of these patients died during this post-OLT period) degrees of liver fibrosis (P = 0.033). The number of patients who developed a histologically demonstrable recurrence at 6 months post-OLT with a degree of fibrosis ≥ 2 after receiving a graft with <30% steatosis versus ≥30% steatosis rose from 10% to 25% (P = 0.306). Of patients who developed a histologically demonstrable recurrence at 12 months post-OLT after receiving a graft with <30% steatosis, 83% and 17% had fibrosis scores < 2 or ≥ 2, respectively. However, 60% and 40% of those receiving a graft with ≥30% steatosis with a histological recurrence showed fibrosis scores < 2 or ≥ 2, respectively (P = 0.035).
Of those patients who developed a histologically demonstrable recurrence during the late post-OLT period after receiving a mildly steatotic graft, 83% and 17% had fibrosis scores < 2 or ≥ 2, respectively. However, 25% and 75% of those patients receiving a moderately or severely steatotic graft had fibrosis scores < 2 or ≥ 2, respectively (P = 0.009).
The overall outcome of patients undergoing OLT for HCV cirrhosis has been widely studied by the most important groups worldwide in liver transplantation. The worse prognosis and increasingly poor outcome of this group of patients have been a matter of debate in the last decade. The limited number of available liver donors has forced the use of marginal livers that do not meet the optimal criteria to become more frequently used for OLT. Different studies have been conducted to determine the influence of marginal donors on the poor evolution in HCV transplanted patients.37, 38 The age and body mass index of the donor, viral load, genotype, living donor liver transplantation, cold and warm ischemia times, number of episodes of acute rejection, immunosuppressive regimens, and some other factors have been reported to modify the overall outcome after OLT for HCV cirrhosis. Unfortunately, the concept of the marginal donor is not the same for all transplant surgeons. In this sense, Melendez and Heaton39 pointed that “the definition of what constitutes a marginal graft will continue to vary between centers until reliable parameters are available for prospectively predicting early graft function.”
Hepatic steatosis is common in the general population, ranging from 6% to 24% upon autopsy examination,40 and is present in about 13% of donor organs.41 Unfortunately, unclear and different conclusions have been stated by different groups about whether the use of steatotic grafts is safe or not for OLT and mainly for HCV recipients. Recent clinical reports have suggested that the use of donor livers with moderate42, 43 and severe steatosis40 for transplantation has no influence on the survival rate or primary nonfunction. Other studies suggest that livers with microvesicular steatosis can be safely used to expand the donor pool.43, 44 However, the use of donor livers with severe fatty infiltration has been reported to be frequently associated with primary hepatic nonfunction after OLT.45 Fatty livers are highly susceptible to cold or warm ischemia injury.46, 47 The ischemia-associated injury may be related to the sinusoid microcirculation disruption caused by the release of intracellular hepatocyte fatty inclusions,48 fatty solidification during cold preservation,45 and sinusoid narrowing that causes blood cell adhesion and Kupffer cell activation.46 On the basis of these findings, there are clinical reports that have definitely found a positive relationship between donor graft steatosis and overall outcome after liver transplantation. Verran et al.49 concluded that the combination of both older donor age and moderate to severe steatosis may adversely impact early allograft survival, stating that the appropriateness of placing a moderately to severely steatotic donor liver into a poor recipient may require consideration. In this study, the rate of allograft loss for HCV recurrence is significantly higher in grafts with >10% steatosis than in those with <10% steatosis (P = 0.058), and so is the 3-month mortality in the subgroups of patients who received a graft with steatosis > 30% (P = 0.02).49 Unfortunately, as the authors commented, “the low number of recipients undergoing OLT with moderately to severely steatotic donor livers along with the many variables that may have had an impact on outcome make firm conclusions difficult.” Salizzoni et al.50 suggested that livers with >15% macrovesicular steatosis may be marginal livers. In that study, the authors pointed out that the risk of graft nonfunction or death of the patient after OLT rises if >15% macrovesicular steatosis is associated with a long ischemia time, high donor age, or HCV positivity in recipients.
The kind of preservation solution used is also a matter of discussion. We routinely used University of Wisconsin solution for liver preservation. Different conclusions are described in the literature about the better results obtained with histidine-tryptophan-ketoglutarate solution51 and Celsior,52, 53 mainly when the cold ischemia time is more than 10 hours.54 In our hospital, we have not introduced the use of these 2 preservation solutions. However, this makes our results comparable to most of the reports appearing in the literature and also avoids an important confusion factor in explaining our results.
According to previous reports, in our cohort of patients, there was a high prevalence of pre-OLT and post-OLT diabetes mellitus among HCV recipients,55–57 but there was not a different distribution among the recipients classified according to donor graft steatosis groups. In our opinion, the mechanism for more frequent and severe viral recurrence must be associated with early colonization of the steatotic graft due to increased susceptibility but not due to long-term diabetes-induced susceptibility. Different studies have addressed the influence of different factors, such as antiviral therapies,58 the number of episodes of rejection,59 immunosuppressive regimens,18 donor age,10, 14 and the molecular virology of HCV60, 61 as well as the genotype and cytomegalovirus coinfection,62 on the progression of hepatic fibrosis after OLT in patients with recurrent HCV infection. Two recent studies have reported that increasing graft steatosis and body weight gain post-OLT do not correlate with early liver fibrosis development.63, 64 Results in both studies are always expressed in terms of post-OLT steatosis evolution, and probably a multivariate analysis of some other marginal donor variables should have been included as well as a comparison of fibrosis evolution in each group of steatotic patients. In that sense, the study from Toniutto et al.64 establishes 10% graft steatosis as the cutoff point, grouping mildly, moderately, and severely steatotic grafts (as defined by Adinolfi et al.26) into 1 group.
The specific contribution of donor graft steatosis to the overall graft outcome and viral recurrence has not been clearly defined. There are no reports that analyze the global influence of progressive donor graft steatosis in HCV recipients versus non-HCV recipients or its effect on fibrosis progression due to increased severity of viral recurrence. Only 1 important recent report has tried to analyze this relationship. Botha et al.65 did not find any association between mild to moderate donor liver steatosis and time to viral recurrence in a cohort of 113 patients who underwent OLT for HCV-related cirrhosis. Furthermore, they could not observe any difference in the time to viral recurrence between patients receiving grade I steatotic livers and those receiving grade II and III steatotic livers, suggesting that “the results of this study support the continued use of mild-moderately steatotic donor organs in the face of an ever-increasing demand, regardless of HCV status.” Unfortunately, the concept of mildly to moderately steatotic donor organs is based on a classification in which steatotic grades 0, I, and II indicate less than 30% macrovesicular fat content and grade III indicates 30% to 45% macrovesicular fat content. Mild to moderate steatosis, as defined by this group, should then include only livers with steatosis lower than 30% as our results suggest. As only 3 patients received grade III grafts and 18 received grade II grafts, the previous statement can be applied only to patients receiving a graft with steatosis lower than 30%, and as a result of the low number of patients, a reliable conclusion can be applied only for grafts with less than 15% liver fatty content (grades 0 and I). These discrepancies can be explained by the different methods used for the classification of the patients according to the liver fat content. Grafts with >30% or >60% steatosis have been used by our team in some patients because of the absence of a routine protocol for donor preretrieval biopsies and on the basis of the judgment of the harvesting surgeon.
In agreement with other studies,37, 66 an extended cold ischemia time (over 12 hours) and donor age (a cutoff point of 55 years) negatively influence the prognosis in HCV recipients. The inclusion of the presence of macrovesicular steatosis, in conjunction with an older donor and prolonged cold ischemia, may act through an increase in liver preservation injury. Future studies may elucidate a relationship between preservation injury and an increase in viral recurrence and liver fibrosis in HCV recipients.
Our study shares some common limitations with the study of Botha et al.,65 such as the lack of pre-OLT viral loads, post-OLT liver biopsies that were event-driven rather than protocol-driven, and the lack of randomization of the patient selection, which allowed only retrospective analysis.
In conclusion, our results show a direct relationship between marginal donors, graft steatosis, and more frequent, more severe, and earlier viral recurrence after OLT for HCV-related cirrhosis. The use of a marginal donor should imply the reduction of the marginal variables, especially the cold ischemia time. The association of donor graft steatosis and marginal variables may increase ischemia-reperfusion injury and favor the susceptibility of the graft to viral recurrence. The frequency and severity of viral recurrence dramatically increase when donor graft steatosis is higher than 30% during OLT in HCV patients. Further multicenter studies and a global consensus may be necessary to finally assess if the use of expanded criteria grafts is safe for HCV-positive recipients and if the organ allocation system needs to be changed for this cohort of patients.