End-stage liver disease associated with chronic infection with hepatitis C virus (HCV) is a common indication of orthotopic liver transplantation (OLT) in the United States. Post-OLT recurrence of HCV is a universal event and may be associated with allograft failure and increased mortality. Indeed, recurrent infection leads to cirrhosis in 10% to 25% of transplant recipients within 5 to 10 years, and once cirrhosis occurs, the 1-year risk of hepatic decompensation reaches 40%.1–3 This emphasizes the need for better understanding of the natural history of post-OLT recurrent HCV and identification of factors associated with rapid fibrosis progression.
Metabolic syndrome (MS), characterized by a constellation of factors including obesity, impaired fasting glucose, hypertension, and dyslipidemia, often occurs after OLT. These metabolic derangements are well-known side effects of immunosuppressant agents used in OLT. Calcineurin inhibitors are associated with the development of hypertension and glucose intolerance, whereas sirolimus may lead to dyslipidemia. In fact, 37% of patients transplanted for chronic HCV infection develop new onset diabetes mellitus (DM) after OLT versus 10% of those transplanted for other indications.4 MS, representing the clinical manifestation of insulin resistance (IR), is associated with chronic HCV infection.5–7 An experimental mouse model transgenic for the HCV core gene has recently been found to be able to induce steatosis, IR, and type II diabetes.8 Moreover, IR measured by the homeostasis model of assessment has been implicated in the development of steatosis and progression of fibrosis in patients with chronic HCV outside the transplant setting.9–11 IR and the resultant hyperinsulinemia may induce hepatic fibrosis in HCV patients by its direct effect on hepatic stellate cells or by altering cytokine production.12
Although the relationship between chronic HCV and the development of IR and MS has been demonstrated in non-OLT patients, the association between post-OLT recurrent HCV and MS remains poorly understood, and the impact of MS on disease progression in patients with recurrent HCV after OLT has not been previously assessed. Post-OLT MS is of particular interest because of its increasing prevalence in the transplant population and its potentially preventable and modifiable nature.
The specific aims of this study were to estimate the prevalence of MS in patients with recurrent HCV after OLT and to assess the association between MS and fibrosis progression in post-OLT recurrent HCV infection with multiple liver biopsy specimens.
ACR, acute cellular rejection; ATP III, Adult Treatment Panel III; BMI, body mass index; CI, confidence interval; CMV, cytomegalovirus; DM, diabetes mellitus; HCV, hepatitis C virus; HDL, high-density lipoprotein; HR, hazard ratio; IR, insulin resistance; MMF, mycophenolate mofetil; MS, metabolic syndrome; OLT, orthotopic liver transplantation; SVR, sustained virological response.
PATIENTS AND METHODS
Using an electronic pathology database, we reviewed the medical records of all adult patients (age > 18 years) who underwent OLT for end-stage liver disease secondary to chronic HCV infection at the Cleveland Clinic between 1998 and 2005. Patients with at least 2 posttransplant liver biopsies were included.
Recurrent hepatitis C was defined by the presence of lobular and portal inflammation in patients with detectable HCV RNA after the exclusion of other causes of histological injury such as graft rejection, biliary disease, vascular complications, and confounding viral infections including cytomegalovirus (CMV).
The presence or absence of post-OLT MS was defined on the basis of the individual patient's characteristics at 1 year post-OLT according to the National Cholesterol Education Program Adult Treatment Panel III (ATP III).13 The ATP III definition requires the presence of any 3 of the following 5 criteria: (1) serum triglycerides > 150 mg/dL or drug treatment for elevated triglycerides; (2) serum high-density lipoprotein (HDL) cholesterol < 40 mg/dL in men or < 50 mg/dL in women or drug treatment for low HDL; (3) blood pressure > 130/85 mm Hg or drug treatment for elevated blood pressure; (4) fasting glucose > 100 mg/dL or drug treatment for elevated blood glucose; and (5) abdominal obesity, which is defined as a waist circumference in men > 102 cm (40 inches) and in women > 88 cm (35 inches). The last trait in the ATP III criteria (waist circumference) was replaced by a body mass index (BMI) greater than 28.8 kg/m2 in both men and women for the purpose of this study. This cutoff value for BMI was equivalent in a regression analysis to a waist circumference of 102 cm in cross-sectional studies of men in the United States and in Europe.14, 15
Patients with no histologic follow-up to assess fibrosis progression post-OLT were excluded. Patients with less than 1 year of follow-up after OLT or those who received liver transplants from a live donor were excluded. We also excluded subjects who were referred for a second OLT or multiple solid organ transplants during the time period from which study patients were identified. Patients with no histologic evidence of recurrent HCV post-OLT were not included in the study.
Demographic (patient's age, gender, race, and donor age) and virologic (HCV genotype and HCV RNA level) information were obtained from prospectively collected data for liver transplant recipients at the Cleveland Clinic. Additional time points at 4 months and at 12 months post-OLT were collected for HCV RNA levels. Other information that was extracted included episodes of acute cellular rejection (ACR) that required corticosteroid treatment and CMV viremia that required antiviral therapy. BMI, fasting serum glucose, triglyceride and HDL, and systolic and diastolic blood pressure at 1 year after OLT were also recorded. Detailed medical histories including the history of hypertension, hyperlipidemia, and DM and their corresponding therapy were obtained form medical records. Additionally, details of antiviral therapy received after liver transplant were obtained from post-OLT HCV treatment–related data sheets filled out prospectively for all liver transplant patients undergoing therapy.
Protocol liver biopsies were performed (day 7, month 4, and yearly) after OLT. Additional biopsies were performed as clinically indicated. In patients treated with antiviral therapy, liver biopsy was performed prior to the start of therapy. All biopsies were read by dedicated liver pathologists. Hepatic fibrosis (stage) was assessed by the Ludwig-Batts scoring system and accordingly assigned with a scale of 0 to 4 (F0: absent, F1: portal fibrosis, F2: periportal fibrosis, F3: bridging fibrosis, F4: cirrhosis).16 Advanced hepatic fibrosis was defined as the presence of stage F3 or F4 fibrosis according to the Ludwig-Batts scoring system.
The first biopsy that showed at least stage 1 fibrosis post-OLT was used for the time-to-progression analysis under the assumption of F0 for all patients at the time of OLT. If there was no evidence of fibrosis during post-OLT follow-up, the last biopsy performed was chosen for the time-to-progression analysis.
The standard immunosuppressant regimen at the time of hospital discharge after OLT consisted of a calcineurin inhibitor (tacrolimus or cyclosporine) and a tapering dose of prednisone. Mycophenolate mofetil (MMF) was added to the regimen when side effects precluded a full therapeutic dose of calcineurin inhibitors. Sirolimus was permitted 4 to 6 months after the surgery, either alone or in combination with either calcineurin inhibitors or MMF. Induction therapy at our institution was started in May 2004 and consisted of 2 doses of interleukin-2 receptor antibodies (basiliximab) on postoperative days 0 and 4. Patients with developed biopsy-proven ACR were treated with a tapering dose of oral or intravenous corticosteroids (an intravenous bolus of 1 g of methylprednisolone followed by a tapering dose over 5 days). During the last 3 years, mild ACR was treated in most HCV patients with calcineurin inhibitor dose adjustment, whereas corticosteroids were reserved for those with a severe rejection episode (rejection activity index > 6/9).
Virological and Laboratory Assays
Quantitative polymerase chain reaction assay (Roche Cobas Amplicor HCV Monitor Test, version 2.0) was performed to detect HCV RNA at 4 months and 1 year after OLT. Prospective surveillance cultures were obtained by protocol, and CMV was isolated from a buffy coat with conventional cell culture methods.17 A blood buffy coat was cultured for CMV at weekly intervals during the first 2 months post-OLT and when clinically indicated. CMV infection in this study was defined as CMV viremia that required treatment with ganciclovir. Fasting serum glucose, plasma triglyceride and HDL, systolic and diastolic blood pressure, and BMI were measured at least twice at 1 year post-OLT, and the average values were used to identify those with MS according to the ATP III criteria.
Descriptive statistics were computed for all factors. These include medians and percentiles for continuous factors and frequencies for categorical factors. In order to assess the association of several baseline factors and liver disease progression, univariable and multivariable proportional hazards analysis was performed. A Kaplan-Meier plot was constructed for visual representation of the (unadjusted) association of MS and progression of fibrosis. Interactions of obesity, MS, and diabetes with time of follow-up were assessed. MS, RNA HCV at 4 months, pulse therapy, and CMV infection were included in a multivariable model. A significance level of 0.05 was considered for all analyses. SAS version 9.1 software (SAS Institute, Cary, NC) was used to perform all analyses.
Patients' Characteristics and Prevalence of MS
A total of 114 patients who had liver transplantation for chronic HCV infection between 1998 and 2005 were identified. Nineteen (16%) subjects were excluded for having 1 or more exclusion factors. In our study population of 95 patients with recurrent HCV after OLT, a total of 409 post-OLT liver biopsies were performed. The median follow-up time from OLT to last available biopsy regardless of when fibrosis observed was 1.6 years (Q25, Q75: 0.99, 2.5 years). Patients who had no evidence of fibrosis during post-OLT follow-up were followed for a median time of 12.8 (8.7, 26.3) months, whereas those who had at least stage 1 fibrosis post-OLT had a median time to progression of 9.5 (5.4, 16.0) months (P = 0.017). Baseline characteristics of this population are listed in Table 1. The median age was 54 years with predominant male gender (84.2%) and Caucasian ethnicity (73.7%). The majority of the patients (81.3%) were infected with HCV genotype 1.
Table 1. Clinical Features of the Study Patients With or Without Fibrosis Progression After Liver Transplantation
All (n = 95)
Progression (n = 68)
No Progression (n = 27)
Abbreviations: ACR, acute cellular rejection; BMI, body mass index; CMV, cytomegalovirus; HCV, hepatitis C virus; OLT, orthotopic liver transplantation; SVR, sustained virological response.
Numbers in parenthesis are range or mean.
Follow-up time for time-to-progression analysis is defined as months from OLT to the first biopsy that showed at least stage 1 fibrosis or months to last biopsy if there was no evidence of fibrosis during post-OLT follow-up.
All patients received a deceased donor liver allograft with a mean donor age of 45 years. The median serum HCV RNA level was 1.94 × 106 IU/mL at 4 months post-OLT and 2.11 × 106 IU/mL at 1 year post-OLT. Twenty-three patients (26.4%) were obese (defined as BMI greater than 30 kg/m2 at 1 year post-OLT) with a median BMI of 27 kg/m2. Additionally, 49 (51.6%) had either glucose intolerance or type II DM.
Immunosuppressive therapy was primarily tacrolimus-based [80/95 (84.2%)], whereas the remaining few patients received either a cyclosporine-based regimen or a sirolimus-based regimen. MMF was used as an adjuvant medication in 32 of 95 (33.7%) patients. Biopsy-proven ACR occurred in 43 of 95 (45.3%) patients, and most were treated with pulse corticosteroids. Chronic rejection was observed in 3 of 95 (3%) subjects. Twenty-one patients (22%) developed symptomatic CMV viremia after OLT requiring ganciclovir therapy.
Antiviral treatment was evaluated as a predictor variable of fibrosis progression only if administrated during the follow-up period from OLT to the first biopsy that showed fibrosis or from OLT to the last biopsy performed if there was no fibrosis progression. Antiviral therapy (regular or pegylated interferon plus ribavirin) was given to 24 (25.3%) patients during post-OLT follow-up. Sustained virological response was achieved in 5 of 24 (21.7%) patients.
The overall median rate of fibrosis progression was 0.08 fibrosis stages per month (Q25, Q75: 0.0, 0.17). The fibrosis progression rate was linear over time in this population with a median fibrosis stage of 1 (Q25, Q75: 0, 2) at 1 year after transplantation. Of 95 patients, 27 (28.4%) had no evidence of fibrosis and 68 (71.6%) had fibrosis progression (any fibrosis during follow-up) during the post-OLT follow-up (Table 1). Of the 27 patients with fibrosis progression, 19 (70.3%) developed advanced fibrosis (F3 or F4) during the follow-up period after OLT.
Complete data for the assessment of MS at 1 year post-OLT were available for 82 of the 95 patients of our cohort (86.3%). In comparison with pre-OLT characteristics, we observed a significant increase in the frequency of hypertension [18/82 (19%) pre-OLT versus 61/82 (64.2%) post-OLT, P = 0.004] and DM [21/82 (22%) pre-OLT versus 49/82 (51.6%) post-OLT, P = 0.0001] after OLT. Forty-one of 82 (50%) patients with recurrent HCV met the ATP III criteria for MS at 1 year after liver transplantation. There was a gender difference in the prevalence of MS, with a slightly higher prevalence of MS in female patients compared to their male counterparts, although the difference did not reach statistical significance (61% versus 47%, P = 0.2). Additionally, there was no association between the development of post-OLT MS and HCV genotype, ethnicity, or type of immunosuppressive treatment regimen.
Predictors of Fibrosis Progression
Univariable and multivariable analyses of factors associated with fibrosis progression in this population are shown in Tables 2 and 3. In univariable analysis (Table 2), a higher HCV viral load at 4 months post-OLT (P < 0.001), CMV infection (P = 0.006), DM (P = 0.046), MS (P = 0.049), and use of antiviral therapy (P < 0.001) were associated with a higher hazard for fibrosis progression after OLT. There was no significant association between fibrosis progression and obesity defined as BMI ≥ 30 (P = 0.65). A Kaplan-Meier plot outlining the association between MS and fibrosis progression is illustrated in Fig. 1.
Table 2. Factors Associated with Fibrosis Progression by Univariable Analysis
Abbreviations: ACR, acute cellular rejection; BMI, body mass index; CI, confidence interval; CMV, cytomegalovirus; HCV, hepatitis C virus; HR, hazard ratio; SVR, sustained virological response.
HCV RNA at 4 months
HCV RNA at 1 year
BMI > 28.8
HCV antiviral therapy
Corticosteroid therapy for ACR
Table 3. Factors Associated with Fibrosis Progression by Multivariable Analysis
The association between fibrosis progression and MS was found to significantly differ with follow-up time (for example, the interaction between MS and follow-up time is statistically significant; P = 0.04). With adjustments for HCV RNA at 4 months post-OLT, CMV infection, and use of corticosteroids, there was evidence to suggest that those with MS and at least 1 year of follow-up had 6.7 (e[ln(HR for MS) + ln(HR for interaction term)] = e[ln(1.06) + ln(6.3)], where HR is the hazard ratio) times the hazards of fibrosis progression in comparison with those without MS. In contrast, MS was not associated with fibrosis progression during the first year after OLT [HR (95% CI): 1.06 (0.49, 2.26); Table 3]. In addition, high HCV RNA at 4 months post-OLT (HR = 1.1, P = 0.004) and the use of corticosteroids to treat acute rejection (HR = 1.9, P = 0.05) remained significantly associated with fibrosis progression.
Survival rates after OLT have substantially improved over the past 2 to 3 decades, although the improvement is mainly due to a reduction in early post-OLT mortality. Although early postoperative mortality in HCV patients undergoing OLT is comparable to that in non-HCV groups, these individuals experience a universal virologic and histological recurrence leading to progressive injury and subsequently decreased survival over the first decade following OLT.1–3, 18 The results of retransplantation for recurrent HCV have also been disappointing, with a 5-year survival rate of less than 50%.19 Identifying modifiable factors promoting rapid fibrosis progression in HCV patients following OLT is of paramount importance and may lead to novel approaches in the management of this population.
Several epidemiological studies have shown that OLT is associated with an increased prevalence of risk factors for cardiovascular disease, including hypertension, DM, obesity, and dyslipidemia.17, 20, 21 However, these metabolic abnormalities have also been linked to fibrosis progression in patients with nonalcoholic fatty liver disease and those with chronic HCV outside the transplantation setting. DM has been described as a risk factor for fibrosis progression and survival after OLT.7, 22 The potential additive role of individual metabolic defects in the setting of MS where multiple abnormalities exist in the same individual has not been assessed in the setting of post-OLT HCV recurrence. Additionally, MS is a constellation of several metabolic variables associated with IR, which has in turn been linked to fibrosis progression in patients with chronic HCV.
In the current study, MS defined by the ATP III criteria was highly prevalent in our patients with recurrent HCV infection following OLT (50%); this level is significantly higher than the prevalence of MS (22%) reported in 8814 adults in the United States participating in the National Health and Nutrition Examination Survey III13 and higher than reported rates of MS (25%) in HCV-infected individuals outside the transplant setting (N. N. Zein et al., manuscript accepted by Clinical Gastroenterology and Hepatology, 2008). Laryea et al.23 recently reported the prevalence of MS in 118 adult liver recipients who had a minimum of 18 months of posttransplantation follow-up. A total of 58% of their patients had MS during the post-OLT period, and more importantly, hepatitis C was found to be a statistically significant variable correlated with MS with an odds ratio of 3.43 (95% CI: 1.90-6.21; P < 0.001). The underlying etiology of MS in this group of patients is unclear and may include viral and nonviral causes. Because HCV has been shown experimentally to induce IR, a higher viral burden after OLT may theoretically have a greater effect on the development of IR after OLT. However, our findings of similar mean levels of HCV RNA in our patients with and without MS at 1 year post-OLT do not support this hypothesis. Alternatively, the immunosuppressive drug regimen may play a role in the development of MS after OLT, especially because several of these medications are associated with an increased risk for individual metabolic complications such as DM, hypertension, or hyperlipidemia.24 However, most of our patients received similar regimens of medications after OLT, and there was no association between MS and any individual agent; this suggests that the immunosuppressive drug regimen alone is unlikely to explain the findings in this study.
The most clinically significant finding of this study is the association between MS and fibrosis progression in patients with recurrent HCV after OLT. The association between MS and fibrosis progression was significant only during a follow-up time that exceeded 1 year after OLT, and this may be due to the time needed for the development of MS after OLT or to the slow rate of fibrosis progression over a period of 1 year. After adjustments for relevant variables, subjects with MS at 1 year post-OLT had 6.7 times the hazards of fibrosis progression in comparison with those without MS. Although MS was independently associated with fibrosis progression, obesity and DM were not. This is consistent with the fact that MS is more reliably associated with risk for cardiovascular disease than its individual components and is almost always associated with measurable IR rather than its individual components.25 Additionally, these observations may explain discrepancies in the medical literature regarding the role of obesity and DM in fibrosis progression in the setting of HCV. Other factors that were independently associated with fibrosis progression in this population included HCV RNA at 4 months post-OLT and steroid therapy for ACR, both of which have been identified in previous studies.18 The association between CMV infection and fibrosis progression was significant only by univariable analysis. Similarly, we found by univariable analysis that patients who received antiviral therapy for HCV were more likely to have fibrosis progression. This association is likely due to selection bias in which candidacy for HCV antiviral therapy after OLT is based on the presence of fibrosis by liver biopsy.
The mechanism of accelerated fibrosis progression in HCV patients after OLT remains unknown. It is believed that IR promotes fibrosis progression through one of several mechanisms, including induction of hepatocyte steatosis, increased tumor necrosis factor alpha production, and impaired expression of peroxisome proliferator-activated receptor-gamma receptors.12 A number of these mechanisms are modifiable, and newer agents developed for the treatment of DM may have to be assessed in future studies for its potential benefit in the setting of recurrent HCV post-OLT.
Several shortcomings of this study can be identified, including the relatively small sample size. However, the findings provide appropriate background for larger prospective studies in which a formal assessment of IR can be made. The use of BMI as an alternate criterion to waist circumference in the definition of MS is another potential shortcoming. However, this issue has been addressed in studies of DM and cardiovascular risk associated with MS. In a prospective study that included 822 nondiabetic subjects (the Insulin Resistance Atherosclerosis Study), the use of BMI in place of waist circumference in the ATP III definition of MS had no impact on diabetes prediction in this population.26, 27 The odds ratio for prediction of DM was 4.14 (95% CI: 2.79-6.16) with the original ATP III definition versus an odds ratio of 4.36 (95% CI: 2.93-6.48) when BMI > 30 kg/m2 was used for both men and women in place of waist circumference in the same population.26 We chose to use a BMI > 28.8 kg/m2 in our study because of 2 cross-sectional studies in men and because it may have a slightly higher sensitivity for identifying those with MS than a higher BMI.14, 28 Finally, although the current study identified a strong epidemiological relationship between MS and fibrosis progression after OLT, future studies should assess the reversibility of MS in this special population and the impact of reversibility on disease progression.
In summary, MS is frequently seen in OLT patients with recurrent HCV and is associated with more rapid fibrosis progression. Given its preventable nature and the simplicity of its assessment, MS could potentially be incorporated into routine post-OLT clinical practice to identify those at greatest risk for poor outcome and in whom therapeutic interventions could be most useful. Future studies are underway to explore the role of several approaches used to reverse IR, including physical activity, weight loss, and insulin-sensitizing agents (metformin and thiazolidinediones) in patients with recurrent HCV after liver transplantation.